Re: [CentOS] Scripting help please....
From: John Doe
> A quick and dirty example (only prints the extra duplicate lines; not the
> original duplicate):
> awk -F: ' { v[$3]=v[$3]+1; if (v[$3]>1) print $0; } ' datafile
Here's the version will the 1st duplicate included:
awk -F: ' { v[$3]=v[$3]+1; if (v[$3] == 1) { f[$3]=$0; } else { if (v[$3] == 2)
print f[$3]; print $0; } } ' datafile
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
Yes it works, but using fqdn as destination, iptables can take too long time to resolve the address. Using nslookup is a better solution, I think. Vinícius Coque On Wed, Oct 28, 2009 at 6:44 PM, Marcus Moeller wrote: > Hi all, > > does it work to define iptables rules with a fqdn as destination > instead of an IP address? Or is it useful to resolve the name first > using e.g. nslookup, writing the result to a variable which is then > used within the -d statement? > > Best Regards > Marcus > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
On 10/29/2009 10:29 AM, Vinicius Coque wrote: >> does it work to define iptables rules with a fqdn as destination >> instead of an IP address? Or is it useful to resolve the name first >> using e.g. nslookup, writing the result to a variable which is then >> used within the -d statement? I guess that depends on what you are trying to achieve, afaik iptables will not hit DNS for each packet, and will only resolve at time of table / policy creation. - KB ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] vnc for non-root
On 10/28/2009 06:35 AM John Doe wrote: > From: ken >> At work I've been asked to set up vnc for a remote user (a vendor >> sysadmin to install 3d party software we've purchased). Of course I'm a >> bit skittish about allowing root access to this. Is there a way to >> configure vnc so that root cannot log in through it...? Or do I have to >> use some other utility to deny root access (e.g., securetty). > > Maybe check the wiki: http://wiki.centos.org/HowTos/VNC-Server > I would guess you just don't give root's vnc password? > > JD That's a better doc than the one I was looking at... got it working now. Thanks much. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] ata errors from dmesg
dmesg is reporting
ata1: translated ATA stat/err 0x51/10 to SCSI SK/ASC/ASCQ 0xb/14/00
ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x10 { SectorIdNotFound }
My HD disk is on /dev/sda1 so I presume this is errors from the CDROM.
Is there a command I can use to not do anything with the CDROM?
It was just there for install. This is an older machine running 4.4 i686 .
Its also remote so I just can't unplug the CDROM.
If there is a command to tell the kernel to ignore the CDROM that
would be great.
Thanks,
jerry
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] vnc for non-root
On 10/28/2009 09:04 AM Neil Aggarwal wrote: > Ken: > > You can set the machine up to use VNC for the > console. > > Then, give the person a normal login which they will > use to login to the machine from the console interface. > Basically, it will be just like they are sitting at > the machine a logging in with a user account. > > I would also require the VNC to be tunneled through > SSH for encryption since VNC does not do that internally. > > Neil Neil, Thanks for the reply. Is it possible to make the ssh connection connecting to a linux box from a windows machine. The only ssh clients we have now for windows are winscp and putty. Realvnc (the Windows vnc client we have) has an encryption option (a box you can click), but it's grayed out (not clickable). Is there some other windows client which has the ssh capability? Thanks again. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] sudoers file
On Oct 29, 2009, at 1:35 AM, vijay shanker wrote: > Thanks guys, > > I have done my changes in the sudoers file. > > what i did is ; added a group with same access as root. > > how i am able to use sudo. but there is a problem. > > my machine is responding very slow for the sudo. It takes almost 3 > minutes to open a small file with command > > sudo vim filename.conf > > i don't think this might be because of the changes. But you can > explain this situation to me. > > To edit sudoers file I used visudo. > > and thanks Majian for that command in vi editor. it was great. 3 minutes is a typical network timeout. Sounds like you don't have dns setup properly and/or nis/winbind or ldap if you use those. -Ross ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] VMware Server 2.0.1 On CentOS 5
It's my understanding that CentOS is a carbon copy of Red Hat - with the exception of the art work. Assuming that this is true, the support matrix for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to locate a CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If I take this route, what should I expect running VMware 2.0.1? Thanks, Gene Poole___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
gene.po...@macys.com wrote: > It's my understanding that CentOS is a carbon copy of Red Hat - with the > exception of the art work. Assuming that this is true, the support > matrix for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to > locate a CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If I > take this route, what should I expect running VMware 2.0.1? My VMware Server 2.0.1 instances all still seem to run normal. I've been running them on 5.3 for quite some time, and this week on 5.4 with no issues. I wouldn't expect any problems. I'm running x86 versions of everything. Regards, Max ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
On Thu, Oct 29, 2009 at 9:29 AM, wrote: > It's my understanding that CentOS is a carbon copy of Red Hat - with the > exception of the art work. Assuming that this is true, the support matrix > for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to locate a > CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If I take this > route, what should I expect running VMware 2.0.1? > > Thanks, > Gene Poole I am running VMware Server 2.0.1 on CentOS 5.2 and 5.3 with no problems. Search this mailing list for info on 5.4, as I think there was a small issue that needed to be worked around. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
Brian Mathis wrote: > I am running VMware Server 2.0.1 on CentOS 5.2 and 5.3 with no > problems. Search this mailing list for info on 5.4, as I think there > was a small issue that needed to be worked around. I believe this was the issue: http://communities.vmware.com/thread/229957 Regards, Max ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
Max Hetrick wrote: > gene.po...@macys.com wrote: >> It's my understanding that CentOS is a carbon copy of Red Hat - with the >> exception of the art work. Assuming that this is true, the support >> matrix for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to >> locate a CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If I >> take this route, what should I expect running VMware 2.0.1? > > My VMware Server 2.0.1 instances all still seem to run normal. I've been > running them on 5.3 for quite some time, and this week on 5.4 with no > issues. I wouldn't expect any problems. I'm running x86 versions of > everything. That's actually pretty strange because the glibc update in 5.4 will break VMware Server 2.0.1. Have you rebooted or restarted vmware since the update? if you haven't, don't until you look up the fix... However, other than this issue which may be fixed in 2.0.2 (I still had a problem but haven't spent much time investigating), the combination works fine. And for the original poster - being 'supported' in vmware just means that they supply a binary kernel module that matches the kernel. You can still run vmware in 'unsupported' combinations but whenever you update the kernel you have to run the script that rebuids the matching module for you before vmware will start. And the 2.0.1 version worked fine with Centos 5.0->5.3. Also, anyone just starting with VMware might want to look at the ESXi version now that it is a free download instead of running the server version. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Switching keyboard language (was: Re: Locales and filenames)
On 10/28/2009 09:10 AM Alfred von Campe wrote: > On Oct 27, 2009, at 19:28, ken wrote: > >> E.g., create a file with vi with just one German/Greek/French word, >> say, >> Έντελέχεια (Entylecheia, an ancient Greek word). If the >> name of the >> file is "nonenglish", then, after you do your save in vim, run the >> shell >> commands >> >> touch temp; mv temp $(cat nonenglish) > > I guess my issue is how these characters get generated in the first > place. By cutting and pasting the word "Έντελέχεια" from > your email into a file on Linux (via the Synergy mouse & keyboard > sharing utility no less), I was able to create a file containing that > word and also named that word and display it correctly with cat and > ls. So UTF-8 encoding appears to work just fine. It's 8-byte > characters in ISO 8859-1 encoding that are causing my problem. > Fortunately, I think I don't have to deal with ISO 8859-1 encodings, > and my problem was self-created by cutting and pasting characters > from the iso_8859-1 man page. > > Now I have a follow up question: so far I've only been able to enter > non-ASCII characters on my Linux system by cutting & pasting; how do > I actually generate any of these characters on a system with a US > keyboard? > > Thanks for all that have helped me solve this problem. > > Alfred There are a lot of keyboard configuration files under /lib/kbd/keymaps/. One of these is loaded at boot-time, probably the one you configured in when you first set up the system. I don't know all the steps you'll need to do-- I've never tried to do what you're doing-- but read the xmodmap manpage and then examine the keycodes in the keymap files mentioned above. For example, mk-utf.map.gz under /lib/kbd/keymaps/i386/qwerty has coding to toggle one keymap to another. IOW, you'd type in one language, hit a couple keys to toggle the keyboard into another language, and then hit another couple/three hotkeys to get back to English... or whichever your home language is. Unless there's some app I don't know about, this is going to be a lot of work, especially if you have to figure out how keymaps work. But work it out and you'll be linux-famous. Document everything. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
Les Mikesell wrote: > That's actually pretty strange because the glibc update in 5.4 will break > VMware > Server 2.0.1. Have you rebooted or restarted vmware since the update? if > you > haven't, don't until you look up the fix... Yeah, I've rebooted my instances. My one instance is my laptop which gets restarted twice a day actually. I've not had any issues at all. Everything has been running great. I'm running this RPM from VMware: VMware-server-2.0.1-156745 So, I'm not sure. Is it something related to x86_64 systems only then? All mine are x86, and I see a lot of references to x86_64 here. Just a thought. Regards, Max ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] problem with 5.4 x86_64 DVD iso?
Hi all, I downloaded the 5.4 x86_64 DVD ISO this morning (CentOS-5.4-x86_64-bin-DVD.iso). However, the md5sum and sha1sums both don't match what's in the md5sum.txt or the release notes (http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.4). I downloaded two ISOs from two different mirrors, RIT and ANL, and both ISOs have the same checksums (match each other), which are different from the published md5sums and sha1sums. Was there a problem with the dissemination of the ISOs to the mirrors? I could understand one mirror having a problem, but two mirrors with the exact same broken image seems like it may be an upstream problem? Anyone have any idea what's up? Thanks, --Joe ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
On Thu, Oct 29, 2009 at 7:01 AM, Max Hetrick wrote: > Les Mikesell wrote: > >> That's actually pretty strange because the glibc update in 5.4 will break >> VMware >> Server 2.0.1. Have you rebooted or restarted vmware since the update? if >> you >> haven't, don't until you look up the fix... > > Yeah, I've rebooted my instances. My one instance is my laptop which > gets restarted twice a day actually. I've not had any issues at all. > Everything has been running great. > > I'm running this RPM from VMware: > > VMware-server-2.0.1-156745 > > So, I'm not sure. Is it something related to x86_64 systems only then? > All mine are x86, and I see a lot of references to x86_64 here. Just a > thought. This issue (including workarounds) is being tracked here: http://bugs.centos.org/view.php?id=3884 Akemi ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Missing package
On Wednesday 28 October 2009 23:42:30 ken wrote: > On 10/28/2009 08:39 AM Anne Wilson wrote: > > The big update caused one of the rare re-starts on my mail server, so I > > saw the startup messages that I had forgotten about. > > > > Oct 28 10:50:04 borg2 python: [3424]: warning: python-dbus not installed. > > Oct 28 10:50:04 borg2 python: hp-systray[3424]: warning: Qt/PyQt 4 > > initialization failed. > > Oct 28 10:50:04 borg2 python: hp-systray[3424]: error: hp-systray > > requires Qt4 GUI and DBus support. Exiting. > > > > I cna't find any package python-dbus. dbus-python is installed, but hp- > > systray doesn't recognise that as the same thing. Is it possible to > > create some sort of link so that when it attempts to call python-dbus it > > gets dbus- python instead? > > > > Anne > > Hmmm. I've got it. > > $ rpm -q dbus-python > dbus-python-0.70-7.el5 > Yes, so have I. But hplip is looking for python.dbus. Anne -- New to KDE4? - get help from http://userbase.kde.org Just found a cool new feature? Add it to UserBase signature.asc Description: This is a digitally signed message part. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
KB is correct--IPTables performs a DNS lookup when it processes the rule. It doesn't slow down to run a DNS lookup for every packet it sees. There are some practical risks to using hostnames, if you're not expecting them, though. If you lose DNS services during startup, your boot will hang for a while trying to resolve those names. Plus, even after it does finish booting, you will be missing the firewall rules that contained the unresolvable names, which may compromise your security to a greater or lesser extent.. Personally, I would avoid using hostnames in iptables startup scripts for these reasons, unless I had some automated notification and fail-safe action for this case, or if I had all the relevant hostnames listed in /etc/hosts or a really persistent local cache, like nscd w/ the 'reload-count infinite' option. On 2009-10-29, Karanbir Singh wrote: > On 10/29/2009 10:29 AM, Vinicius Coque wrote: >>> does it work to define iptables rules with a fqdn as destination >>> instead of an IP address? Or is it useful to resolve the name first >>> using e.g. nslookup, writing the result to a variable which is then >>> used within the -d statement? > > I guess that depends on what you are trying to achieve, afaik iptables > will not hit DNS for each packet, and will only resolve at time of table > / policy creation. > > - KB > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > -- Ryan B. Lynch ryan.b.ly...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] vnc for non-root
Ken: > Is it possible to make the ssh connection > connecting to a linux box from a windows machine. I use cygwin ssh to connect CentOS servers from my windows laptop. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
Max Hetrick wrote: > Les Mikesell wrote: > >> That's actually pretty strange because the glibc update in 5.4 will break >> VMware >> Server 2.0.1. Have you rebooted or restarted vmware since the update? if >> you >> haven't, don't until you look up the fix... > > Yeah, I've rebooted my instances. My one instance is my laptop which > gets restarted twice a day actually. I've not had any issues at all. > Everything has been running great. > > I'm running this RPM from VMware: > > VMware-server-2.0.1-156745 > > So, I'm not sure. Is it something related to x86_64 systems only then? > All mine are x86, and I see a lot of references to x86_64 here. Just a > thought. No, I have it on an x86 box and had to use the workaround here: http://communities.vmware.com/message/1364852 /lib/libc-2.5.so is actually still available after the upgrade so you don't have to copy it from another system - it just isn't the target of the libc.so.6 symlink anymore. Someone reported different behavior on core duo's vs. xeons which is probably what you are seeing, but eventually both crashed. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
On Thu, Oct 29, 2009 at 10:58, Ryan Lynch wrote: > KB is correct--IPTables performs a DNS lookup when it processes the > rule. It doesn't slow down to run a DNS lookup for every packet it > sees. > > There are some practical risks to using hostnames, if you're not > expecting them, though. If you lose DNS services during startup, your > boot will hang for a while trying to resolve those names. Plus, even > after it does finish booting, you will be missing the firewall rules > that contained the unresolvable names, which may compromise your > security to a greater or lesser extent.. > > Personally, I would avoid using hostnames in iptables startup scripts > for these reasons, unless I had some automated notification and > fail-safe action for this case, or if I had all the relevant hostnames > listed in /etc/hosts or a really persistent local cache, like nscd w/ > the 'reload-count infinite' option. > > > On 2009-10-29, Karanbir Singh wrote: >> On 10/29/2009 10:29 AM, Vinicius Coque wrote: does it work to define iptables rules with a fqdn as destination instead of an IP address? Or is it useful to resolve the name first using e.g. nslookup, writing the result to a variable which is then used within the -d statement? >> >> I guess that depends on what you are trying to achieve, afaik iptables >> will not hit DNS for each packet, and will only resolve at time of table >> / policy creation. BTW, sorry for the top-posting. The gmail client for BlackBerry seems to have been designed in the spirit of "Freedom means not having to make a choice". -Ryan ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] sudoers file
No Ross, This is the irony; i am working on the same machine. There is no network in between Regards, Vijay Shanker Dubey Ph: +91-9818311884 On Thu, Oct 29, 2009 at 6:40 PM, Ross Walker wrote: > On Oct 29, 2009, at 1:35 AM, vijay shanker wrote: > > > Thanks guys, > > > > I have done my changes in the sudoers file. > > > > what i did is ; added a group with same access as root. > > > > how i am able to use sudo. but there is a problem. > > > > my machine is responding very slow for the sudo. It takes almost 3 > > minutes to open a small file with command > > > > sudo vim filename.conf > > > > i don't think this might be because of the changes. But you can > > explain this situation to me. > > > > To edit sudoers file I used visudo. > > > > and thanks Majian for that command in vi editor. it was great. > > 3 minutes is a typical network timeout. > > Sounds like you don't have dns setup properly and/or nis/winbind or > ldap if you use those. > > -Ross > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] vnc for non-root
Neil Aggarwal wrote: > Ken: > >> Is it possible to make the ssh connection >> connecting to a linux box from a windows machine. > > I use cygwin ssh to connect CentOS servers from my windows > laptop. Or you could do it with putty. But, if you need more than occasional remote GUI access you should probably look at freenx/NX. It is much more responsive than vnc and runs over ssh by default. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] sudoers file
vijay shanker schrieb am 29.10.2009 16:24:54: > No Ross, > This is the irony; i am working on the same machine. There is no > network in between Install strace, then run sudo via strace and look which syscall is causing the hangs. As always the manpage is your friend. As a sidenote: I don't know if you're aware of it but allowing vi/vim via sudo gives you full access as the user (in your case root). This is true for all programms with the possibility to break out into a shell. You may want to read sudo(8) and sudoers(5) and watch out for noexec. Frank. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
Les Mikesell wrote: > No, I have it on an x86 box and had to use the workaround here: > http://communities.vmware.com/message/1364852 > /lib/libc-2.5.so is actually still available after the upgrade so you > don't have to copy it from another system - it just isn't the target of > the libc.so.6 symlink anymore. > > Someone reported different behavior on core duo's vs. xeons which is > probably what you are seeing, but eventually both crashed. Ok, thanks! I'll keep my eye out. These machines aren't business critical, but I'll definitely keep a watch out. Regards, Max ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] problem with 5.4 x86_64 DVD iso?
From: "Greenseid, Joseph M (IS)" >I downloaded the 5.4 x86_64 DVD ISO this morning >(CentOS-5.4-x86_64-bin-DVD.iso). However, the md5sum and sha1sums both don't >match what's in the md5sum.txt or the release notes >(http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.4). I downloaded two >ISOs from two different mirrors, RIT and ANL, and both ISOs have the same >checksums (match each other), which are different from the published md5sums >and sha1sums. Same here... Mines are: 071e18754c2fb066c526672f9aea0515 CentOS-5.4-i386-bin-DVD.iso 92cab2977a58ce422130e5c655dc8513 CentOS-5.4-x86_64-bin-DVD.iso The md5sum.txt files on the mirrors still contain these... JD ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Infrastructure HELP!
Hi Everyone, I started a social networking site and I am getting ready to go live next week. I have limited fund as I am not funded yet (although hopefully soon). I have an extra Mac Pro (I know, how can I possibly call a Mac Pro dual 2.8 quad core with 16gb RAM extra). So Mac Pro quad Core, 16gb RAM, 4 x 1tb RAID level Seagate drives. I was going to load CentOS 5.4 on it, web, mysql etc, etc. This is really the only box I have that would handle the site it it takes off and then I would need to add more hardware and most hosting to RackSpace or something. My question is about initial setup. The 4 x 1TB drives. How to set this up for I have some protection? RAID 0+1? (striped + mirrored) I would end up with 2TB useable space. RAID 5? so what one is a hot spare? 3TB useable space? What about striping the 4 1TB and mirroring that to a 4 x 1tb External drive enclosure? I want to be sure the data is protected, but machine resources and money are limited. Thoughts are appreciated! ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
gene.po...@macys.com wrote: > It's my understanding that CentOS is a carbon copy of Red Hat - with > the exception of the art work. Assuming that this is true, the support > matrix for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to > locate a CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If > I take this route, what should I expect running VMware 2.0.1? my experience with VMWare Server 2.0.x on x86_64 platforms is abysmal. its totally unstable and unusable, on both Linux (RHEL 5.x) and Windows Vista 64bit, even if you are just running 32bit VMs. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
If you want speed, use RAID 10. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime > -Original Message- > From: centos-boun...@centos.org > [mailto:centos-boun...@centos.org] On Behalf Of ML > Sent: Thursday, October 29, 2009 12:57 PM > To: CentOS mailing list > Subject: [CentOS] Infrastructure HELP! > > Hi Everyone, > > I started a social networking site and I am getting ready to go live > next week. I have limited fund as I am not funded yet (although > hopefully soon). > > I have an extra Mac Pro (I know, how can I possibly call a Mac Pro > dual 2.8 quad core with 16gb RAM extra). So Mac Pro quad Core, 16gb > RAM, 4 x 1tb RAID level Seagate drives. I was going to load > CentOS 5.4 > on it, web, mysql etc, etc. This is really the only box I have that > would handle the site it it takes off and then I would need to add > more hardware and most hosting to RackSpace or something. > > My question is about initial setup. The 4 x 1TB drives. How to set > this up for I have some protection? > > RAID 0+1? (striped + mirrored) I would end up with 2TB useable space. > > RAID 5? so what one is a hot spare? 3TB useable space? > > What about striping the 4 1TB and mirroring that to a 4 x 1tb > External > drive enclosure? > > I want to be sure the data is protected, but machine resources and > money are limited. > > Thoughts are appreciated! > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Neil, Can you explain how RAID 10 would work with 4 x 1tb drives? Should I just get 2 x 2tb drives and mirror? I probably dont need 4 x 1tb drives to start, maybe even 2 x 2tb. I can't image this growing faster than I can get money to add more equipment, move to Co-Lo, etc. On Oct 29, 2009, at 10:59 AM, Neil Aggarwal wrote: > > If you want speed, use RAID 10. > > Neil > > -- > Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com > CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated > 64bit CPU > 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero > downtime > >> -Original Message- >> From: centos-boun...@centos.org >> [mailto:centos-boun...@centos.org] On Behalf Of ML >> Sent: Thursday, October 29, 2009 12:57 PM >> To: CentOS mailing list >> Subject: [CentOS] Infrastructure HELP! >> >> Hi Everyone, >> >> I started a social networking site and I am getting ready to go live >> next week. I have limited fund as I am not funded yet (although >> hopefully soon). >> >> I have an extra Mac Pro (I know, how can I possibly call a Mac Pro >> dual 2.8 quad core with 16gb RAM extra). So Mac Pro quad Core, 16gb >> RAM, 4 x 1tb RAID level Seagate drives. I was going to load >> CentOS 5.4 >> on it, web, mysql etc, etc. This is really the only box I have that >> would handle the site it it takes off and then I would need to add >> more hardware and most hosting to RackSpace or something. >> >> My question is about initial setup. The 4 x 1TB drives. How to set >> this up for I have some protection? >> >> RAID 0+1? (striped + mirrored) I would end up with 2TB useable space. >> >> RAID 5? so what one is a hot spare? 3TB useable space? >> >> What about striping the 4 1TB and mirroring that to a 4 x 1tb >> External >> drive enclosure? >> >> I want to be sure the data is protected, but machine resources and >> money are limited. >> >> Thoughts are appreciated! >> ___ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
> However, other than this issue which may be fixed in 2.0.2 (I still > had a problem but haven't spent much time investigating), the > combination works fine. The upgrade to VMware Server 2.0.2 on 5.4 has not gone well for me. The XP shutdown command hangs. XP Task Manager then shows nothing running any significant cpu load then totally hangs the guest and a forced reboot is required. The forced reboot generates a "Communication Error" . Restarting vmware fails until all processes related to vmware are killed, config.pl is re-run and the server restarted. Upgrading Vmware tools finally succeeded after several tries with only cryptic messages that the "Installation failed". I haven't tried a guest shutdown since the tools were fixed so I don't know if this is or is not related. Not a trace of information in any of the logs. Will pursue this further tonight. Server 2.0.1 and it's predecessors have run flawlessly for me the last few years. In addition, my tea is cold and my little toe hurts. There - I've done all my whining for the day. benm ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
RAID 10 is striping across mirrored drives. So, if you have 4 x 1TB drives, think of it as two separate 1 TB volumes. The system will write half your data to volume A and the other half to volume B. The data in volume A and B do not overlap. Now, each volume is composed of a mirrored set of drives. Anything written to volume A is actually stored on two drives. Anything written to volume B is actually stores on the other two drives. Does this make sense? Let me know if you need any more explanation. Also, when you move to a hosted solution, I would appreciate your considering my company for it. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime > -Original Message- > From: centos-boun...@centos.org > [mailto:centos-boun...@centos.org] On Behalf Of ML > Sent: Thursday, October 29, 2009 1:03 PM > To: CentOS mailing list > Subject: Re: [CentOS] Infrastructure HELP! > > Neil, > Can you explain how RAID 10 would work with 4 x 1tb drives? > > Should I just get 2 x 2tb drives and mirror? > > I probably dont need 4 x 1tb drives to start, maybe even 2 x 2tb. I > can't image this growing faster than I can get money to add more > equipment, move to Co-Lo, etc. > > On Oct 29, 2009, at 10:59 AM, Neil Aggarwal wrote: > > > > > If you want speed, use RAID 10. > > > > Neil > > > > -- > > Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com > > CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated > > 64bit CPU > > 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero > > downtime > > > >> -Original Message- > >> From: centos-boun...@centos.org > >> [mailto:centos-boun...@centos.org] On Behalf Of ML > >> Sent: Thursday, October 29, 2009 12:57 PM > >> To: CentOS mailing list > >> Subject: [CentOS] Infrastructure HELP! > >> > >> Hi Everyone, > >> > >> I started a social networking site and I am getting ready > to go live > >> next week. I have limited fund as I am not funded yet (although > >> hopefully soon). > >> > >> I have an extra Mac Pro (I know, how can I possibly call a Mac Pro > >> dual 2.8 quad core with 16gb RAM extra). So Mac Pro quad Core, 16gb > >> RAM, 4 x 1tb RAID level Seagate drives. I was going to load > >> CentOS 5.4 > >> on it, web, mysql etc, etc. This is really the only box I have that > >> would handle the site it it takes off and then I would need to add > >> more hardware and most hosting to RackSpace or something. > >> > >> My question is about initial setup. The 4 x 1TB drives. How to set > >> this up for I have some protection? > >> > >> RAID 0+1? (striped + mirrored) I would end up with 2TB > useable space. > >> > >> RAID 5? so what one is a hot spare? 3TB useable space? > >> > >> What about striping the 4 1TB and mirroring that to a 4 x 1tb > >> External > >> drive enclosure? > >> > >> I want to be sure the data is protected, but machine resources and > >> money are limited. > >> > >> Thoughts are appreciated! > >> ___ > >> CentOS mailing list > >> CentOS@centos.org > >> http://lists.centos.org/mailman/listinfo/centos > > > > ___ > > CentOS mailing list > > CentOS@centos.org > > http://lists.centos.org/mailman/listinfo/centos > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
On Thu, 29 Oct 2009, ML wrote: > My question is about initial setup. The 4 x 1TB drives. How to set > this up for I have some protection? > > RAID 0+1? (striped + mirrored) I would end up with 2TB useable space. > > RAID 5? so what one is a hot spare? 3TB useable space? > > What about striping the 4 1TB and mirroring that to a 4 x 1tb External > drive enclosure? Plan for drives going out but keeping the site operational. Also remember that RAID isn't a backup. RAID 5 is enticing because you get more usable space out of your drives, but you have to be sure you'll only lose one drive at a time and can get a replacement drive in there and sync'ed up before you lose a 2nd one... If the drives were made by the same manufacturer and were bought at about the same time you might easily lose 2 or more, blowing up your array. Go with RAID 1+0 (mirroring + striping) and you can potentially lose up to 1/2 of your drives and keep running. That's assuming you lose the correct 1/2... If you lose two drives in the same mirror you still go down. Combine the above with a good backup strategy. Backup software has been discussed on this list quite recently so look back a few days or weeks and you'll find some good links. Practice and document doing restores so that you know how to do them quickly without error when the pressure's on. Some of the backup software names/links I gleaned from this list recently: www.mondorescue.org www.nongnu.org/duplicity (de-dupe, S3) www.nongnu.org/storebackup (de-dupe) amanda backuppc (sourceforge) (rpm in epel) www.backula.org (de-dupe) -- Curt Mills, WE7U hacker at fluke dot com Senior Methods Engineer/SysAdmin "Lotto: A tax on people who are bad at math." -- unknown "Windows: Microsoft's tax on computer illiterates." -- WE7U "The world DOES revolve around me: I picked the coordinate system!" Please be advised that this email may contain confidential information. If you are not the intended recipient, please do not read, copy or re-transmit this email. If you have received this email in error, please notify us by email by replying to the sender and by telephone (call us collect at +1 202-828-0850) and delete this message and any attachments. Thank you in advance for your cooperation and assistance. In addition, Danaher and its subsidiaries disclaim that the content of this email constitutes an offer to enter into, or the acceptance of, any contract or agreement or any amendment thereto; provided that the foregoing disclaimer does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment to this email. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] resolv.conf rewritten every reboot. How to figure out who and why?
The consensus of the list seemed to be that I should change the PEERDNS
variable. It seems not to be working. The machine rebooted yesterday,
/etc/resolv.conf got rewritten again. And yet:
find /etc/sysconfig/ -type f -exec grep -iH 'peerdns=' {} \;
/etc/sysconfig/network-scripts/ifcfg-eth0:PEERDNS=no
/etc/sysconfig/network-scripts/ifcfg-eth0.old:PEERDNS=yes
/etc/sysconfig/networking/profiles/default/ifcfg-eth0:PEERDNS=no
/etc/sysconfig/networking/devices/ifcfg-eth0:PEERDNS=no
# chkconfig --list|grep Net
NetworkManager 0:off 1:off 2:off 3:off 4:off 5:off 6:off
# /etc/init.d/NetworkManager status
NetworkManager is stopped
# find /etc/sysconfig/ -type f -exec grep -iH 'bootproto=' {} \;
/etc/sysconfig/network-scripts/ifcfg-eth0:BOOTPROTO=none
/etc/sysconfig/network-scripts/ifcfg-eth0.old:BOOTPROTO=none
/etc/sysconfig/network-scripts/ifup-eth:BOOTPROTO=bootp
/etc/sysconfig/networking/profiles/default/ifcfg-eth0:BOOTPROTO=none
/etc/sysconfig/networking/devices/ifcfg-eth0:BOOTPROTO=none
# cat /etc/resolv.conf
; generated by /sbin/dhclient-script
search soest.hawaii.edu
nameserver 1...
nameserver 1...
So, it's not PEERDNS, not DHCP, not NetworkManager. Why is dhclient-script
even being run?
Maybe I should disable selinux altogether, instead of just making it
permissive? I think I'll try that.
# grep resol /sbin/dhclient-script
make_resolv_conf() {
save_previous /etc/resolv.conf
change_resolv_conf $rscf
for resolvfile in /etc/resolv.conf* ; do
/sbin/restorecon $resolvfile >/dev/null 2>&1
make_resolv_conf
if [ -f /etc/resolv.conf.predhclient.$interface ]; then
change_resolv_conf /etc/resolv.conf.predhclient.$interface
rm -f /etc/resolv.conf.predhclient.$interface
[r...@lee1 ~]# getenforce
Permissive
How badly would things blow up if I did this?
chmod a-w /etc/resolv.conf
mahalo,
Dave
On Thu, Oct 8, 2009 at 7:41 AM, Dave
> wrote:
> My machine has a static IP, with dhcp and IPv6 disabled. Every time I
> reboot, some process rewrites /etc/resolv.conf, including a comment
> about dhcpclient. The only package I have installed that shows up in
> "rpm -qa|grep -i dhcp" is dhcpv6-client-1.0.10-16.el5, and nothing in
> there is named dhcpclient.
>
> I'd like to figure out what software is rewriting this file and why.
> man 5 resolv.conf and man resolver are unhelpful in this case. rpm
> reports /etc/resolv.conf is not owned by any package.
>
> At this point, I am as (or more) interested in pointers regarding how
> to find the answer as I am in the actual answer. Please teach me to
> fish.
>
> mahalo,
> Dave
>
--
———-
Q: Why should this email be 5 sentences or less?
A: http://five.sentenc.es
IPRC-help FAQ: https://wailua/wiki/index.php/Faq
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] vnc for non-root
On 10/29/2009 11:29 AM Les Mikesell wrote: > Neil Aggarwal wrote: >> Ken: >> >>> Is it possible to make the ssh connection >>> connecting to a linux box from a windows machine. >> I use cygwin ssh to connect CentOS servers from my windows >> laptop. > > Or you could do it with putty. But, if you need more than occasional > remote GUI access you should probably look at freenx/NX. It is much > more responsive than vnc and runs over ssh by default. > Thanks, Mike and Neil and everyone, for your help and suggestions. I've got some options to check out. :) Best to all, ken ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] problem with 5.4 x86_64 DVD iso?
Turned out to be a download problem, not a mirror problem. --Joe From: centos-boun...@centos.org on behalf of Greenseid, Joseph M (IS) Sent: Thu 10/29/2009 10:46 AM To: CentOS Mailing List Subject: [CentOS] problem with 5.4 x86_64 DVD iso? Hi all, I downloaded the 5.4 x86_64 DVD ISO this morning (CentOS-5.4-x86_64-bin-DVD.iso). However, the md5sum and sha1sums both don't match what's in the md5sum.txt or the release notes (http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.4). I downloaded two ISOs from two different mirrors, RIT and ANL, and both ISOs have the same checksums (match each other), which are different from the published md5sums and sha1sums. Was there a problem with the dissemination of the ISOs to the mirrors? I could understand one mirror having a problem, but two mirrors with the exact same broken image seems like it may be an upstream problem? Anyone have any idea what's up? Thanks, --Joe ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Curt Mills wrote: > On Thu, 29 Oct 2009, ML wrote: > > >> My question is about initial setup. The 4 x 1TB drives. How to set >> this up for I have some protection? >> >> RAID 0+1? (striped + mirrored) I would end up with 2TB useable space. >> >> RAID 5? so what one is a hot spare? 3TB useable space? >> >> What about striping the 4 1TB and mirroring that to a 4 x 1tb External >> drive enclosure? >> > > Plan for drives going out but keeping the site operational. Also > remember that RAID isn't a backup. > > RAID 5 is enticing because you get more usable space out of your > drives, but you have to be sure you'll only lose one drive at a time > and can get a replacement drive in there and sync'ed up before you > lose a 2nd one... If the drives were made by the same manufacturer > and were bought at about the same time you might easily lose 2 or > more, blowing up your array. > > Right. The problem with 1TB drives (and really, any modern drive with decent capacity) is that you're vulnerable until that array finishes rebuilding-- a process that's taking longer and longer. I'd go with the RAID10 solution that someone previously posited. By the time you outgrow that, you should really plan for a SAN/NAS solution anyway... -- Corey / KB1JWQ ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] printers... (HP)
I can't remember if I asked here - I've got more than one printer, HP, laser, builtin JetDirect. When I go there in a browser, I see the basic thing, but I can't get to the networking page. It says that it wants a password, no user name. It doesn't accept anything. Last week, a co-worker reset one to factory defaults... and it comes up with the same thing. Looking around on the Web, I don't see anything about a default password. Clues for the poor? mark ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
On Oct 29, 2009, at 11:11 AM, Neil Aggarwal wrote: > RAID 10 is striping across mirrored drives. > > So, if you have 4 x 1TB drives, think of it as two separate > 1 TB volumes. The system will write half your data to > volume A and the other half to volume B. The data in volume > A and B do not overlap. > > Now, each volume is composed of a mirrored set of drives. > Anything written to volume A is actually stored on two drives. > Anything written to volume B is actually stores on the other two > drives. > > Does this make sense? > Let me know if you need any more explanation. No it makes sense. I am contemplating if I really need 4 x 1tb in this system. I mean how much space with some photo's, web pages and MySQL take up if there are 5,000 subscribers to start up? Would 2 x 1TB enterprise drives be enough mirrored? > Also, when you move to a hosted solution, I would appreciate > your considering my company for it. Sure, I will be doing a lot of research on that for sure. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Benefits of VMWare 2.0 over 1.x?
I've been running VMWare Server 1.x for some time now, currently on 1.08, and I've been pretty satisfied with it. I was wondering if any of you fellow VMWare users are seeing any significant benefit to moving to the 2.0 release. I'm running CentOS 5.4 with Linux 2.6.18-164.el5 #1 SMP x86_64 on an AMD 7750 64x2 with 4GB of RAM Thanks. mhr ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
ML schrieb: > Hi Everyone, > > I want to be sure the data is protected, but machine resources and > money are limited. > Why don't you rent a VPS for the time being and rsync the file+data to your MacPro, where you can use TimeMachine to create further backups? Rainer ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> Would 2 x 1TB enterprise drives be enough mirrored? Are you going to to software RAID1 or hardware? I find software RAID1 bogs down for intensive database applications. NOTE: Host based RAID is the same as software RAID. You will need an actual external RAID card like one from Areca. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Benefits of VMWare 2.0 over 1.x?
MHR wrote: > I've been running VMWare Server 1.x for some time now, currently on > 1.08, and I've been pretty satisfied with it. > > I was wondering if any of you fellow VMWare users are seeing any > significant benefit to moving to the 2.0 release. > > I'm running CentOS 5.4 with Linux 2.6.18-164.el5 #1 SMP x86_64 on an > AMD 7750 64x2 with 4GB of RAM For me at least 2.x supports the newer kernels, I haven't tried vmware server on centos but the drivers wouldn't build for me on Debian 5.0(even tried some 3rd party patches to no avail). If it weren't for that then I'd probably still be on 1.x. The one thing I do miss about 1.x is the remote console, the new remote console in 2.x is web-based, which is kind of annoying. nate ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] printers... (HP)
On Thursday 29 October 2009 18:35:56 m.r...@5-cent.us wrote: > I can't remember if I asked here - I've got more than one printer, HP, > laser, builtin JetDirect. When I go there in a browser, I see the basic > thing, but I can't get to the networking page. It says that it wants a > password, no user name. It doesn't accept anything. Last week, a co-worker > reset one to factory defaults... and it comes up with the same thing. > Looking around on the Web, I don't see anything about a default password. > > Clues for the poor? > Have you tried installing hplip? You'll have to check whether the CentOS package is late enough to support your printers, and you may have to download a ppd file, but I find it works very well here. I have a mfp, JetDirect, connected by wireless, and a Deskjet 990CXi by usb to my laptop. I've also used the Deskjet on another box connected to the LAN. The setup tool sees them both, and providing it can find the ppd it's really easy. (IIRC the CentOS package didn't support my mfp, but just pointing it to the .ppd was sufficient to get it working.) Anne -- New to KDE4? - get help from http://userbase.kde.org Just found a cool new feature? Add it to UserBase signature.asc Description: This is a digitally signed message part. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Benefits of VMWare 2.0 over 1.x?
On Thu, 29 Oct 2009, nate wrote: > MHR wrote: >> I've been running VMWare Server 1.x for some time now, currently on >> 1.08, and I've been pretty satisfied with it. >> >> I was wondering if any of you fellow VMWare users are seeing any >> significant benefit to moving to the 2.0 release. >> >> I'm running CentOS 5.4 with Linux 2.6.18-164.el5 #1 SMP x86_64 on an >> AMD 7750 64x2 with 4GB of RAM > Much more flexibilty in setting up networking, etc. Agree that the lack of a client is irritating.. -- Jim Wildman, CISSP, RHCE j...@rossberry.com http://www.rossberry.com "Society in every state is a blessing, but Government, even in its best state, is a necessary evil; in its worst state, an intolerable one." Thomas Paine ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
>> Would 2 x 1TB enterprise drives be enough mirrored? > > Are you going to to software RAID1 or hardware? > > I find software RAID1 bogs down for intensive database > applications. If I can find a cheap enough RAID Card that wil fit the Mac Pro, I would do hardware RAID, but Apple wants $699 for theirs... ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] resolv.conf rewritten every reboot. How to figure out who and why?
> /etc/sysconfig/network-scripts/ifup-eth: BOOTPROTO=bootp
> So, it's not PEERDNS, not DHCP, not NetworkManager. Why is dhclient-script
> even being run?
BOOTPROTO=bootp is triggering it.
In /etc/sysconfig/network-scripts/ifup-eth:
if [ "${BOOTPROTO}" = "bootp" -o "${BOOTPROTO}" = "dhcp" ]; then
DYNCONFIG=true
fi
Then, later on in that script:
if [ -n "${DYNCONFIG}" -a -x /sbin/dhclient ]; then
do a bunch of stuff related to dynamically assigned addresses and names.
And finally:
if /sbin/dhclient ${DHCLIENTARGS} ${DEVICE} ; then
echo $" done."
else
> Maybe I should disable selinux altogether, instead of just making it
> permissive? I think I'll try that.
selinux shouldn't have anything to do with the resolv.conf file being
rewritten -- unless it's set to enforcing and something that isn't
allowed to do so is trying to rewrite the file.
HTH,
M
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
If you want performance stick with RAID 10. In general the more drives (spindles) the faster the array. The Western Digital RE3 500 GB drives are a good deal. You should be able to get 4 of those in the low $200s. In RAID 10 this would give you better performance than 2 x 1TB in RAID 1. Ryan On Thu, Oct 29, 2009 at 2:36 PM, ML wrote: > > On Oct 29, 2009, at 11:11 AM, Neil Aggarwal wrote: > >> RAID 10 is striping across mirrored drives. >> >> So, if you have 4 x 1TB drives, think of it as two separate >> 1 TB volumes. The system will write half your data to >> volume A and the other half to volume B. The data in volume >> A and B do not overlap. >> >> Now, each volume is composed of a mirrored set of drives. >> Anything written to volume A is actually stored on two drives. >> Anything written to volume B is actually stores on the other two >> drives. >> >> Does this make sense? >> Let me know if you need any more explanation. > > No it makes sense. > > I am contemplating if I really need 4 x 1tb in this system. I mean how > much space with some photo's, web pages and MySQL take up if there are > 5,000 subscribers to start up? > > Would 2 x 1TB enterprise drives be enough mirrored? > >> Also, when you move to a hosted solution, I would appreciate >> your considering my company for it. > > Sure, I will be doing a lot of research on that for sure. > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> If I can find a cheap enough RAID Card that wil fit the Mac Pro, I > would do hardware RAID, but Apple wants $699 for theirs... $699 is pretty steep, but RAID cards are not cheap. They are worth it for performance though. If you don't need absolute performance, software RAID will work. Give with multiple smaller drives instead of two larger ones. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
On Thu, Oct 29, 2009 at 02:07:22PM -0500, Neil Aggarwal wrote: > > If I can find a cheap enough RAID Card that wil fit the Mac Pro, I > > would do hardware RAID, but Apple wants $699 for theirs... > > $699 is pretty steep, but RAID cards are not cheap. > They are worth it for performance though. > > If you don't need absolute performance, software RAID will > work. > > Give with multiple smaller drives instead of two larger > ones. Agreed on this one. The bigger drives are great, but even in a mirror rebuild scenario the time to re-sync just gets longer and longer. More risk! $700, eesh. You can get some nice Areca cards for much less than that. Ray ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] VMware Server 2.0.1 On CentOS 5
John R Pierce wrote: >> It's my understanding that CentOS is a carbon copy of Red Hat - with >> the exception of the art work. Assuming that this is true, the support >> matrix for VMware Server 2.0.1 states Red Hat 5.1. I cannot seem to >> locate a CentOS 5.1 x86-64 copy. I can get a copy of 5.3 and 5.4. If >> I take this route, what should I expect running VMware 2.0.1? > > my experience with VMWare Server 2.0.x on x86_64 platforms is abysmal. > its totally unstable and unusable, on both Linux (RHEL 5.x) and Windows > Vista 64bit, even if you are just running 32bit VMs. I don't like the change to a web based console in the 2.x series so I'm still mostly using 1.x versions (and I've had some things running under Centos 3.x/VMware 1.x for years with no issues). But, I hadn't seen any real problems with the 2.0.1 version before the glibc change in the 5.4 upgrade. What kind of trouble did you have? -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Ryan, > If you want performance stick with RAID 10. In general the more drives > (spindles) the faster the array. The Western Digital RE3 500 GB drives > are a good deal. You should be able to get 4 of those in the low > $200s. In RAID 10 this would give you better performance than 2 x 1TB > in RAID 1. > They are like $89.99 a piece on NewEgg. I have a friend that has 1 x 1TB Seagate Raid level drives he will sell me for $100 each. Is software RAID 10 decent performance? -Jason > Ryan > > On Thu, Oct 29, 2009 at 2:36 PM, ML > wrote: >> >> On Oct 29, 2009, at 11:11 AM, Neil Aggarwal wrote: >> >>> RAID 10 is striping across mirrored drives. >>> >>> So, if you have 4 x 1TB drives, think of it as two separate >>> 1 TB volumes. The system will write half your data to >>> volume A and the other half to volume B. The data in volume >>> A and B do not overlap. >>> >>> Now, each volume is composed of a mirrored set of drives. >>> Anything written to volume A is actually stored on two drives. >>> Anything written to volume B is actually stores on the other two >>> drives. >>> >>> Does this make sense? >>> Let me know if you need any more explanation. >> >> No it makes sense. >> >> I am contemplating if I really need 4 x 1tb in this system. I mean >> how >> much space with some photo's, web pages and MySQL take up if there >> are >> 5,000 subscribers to start up? >> >> Would 2 x 1TB enterprise drives be enough mirrored? >> >>> Also, when you move to a hosted solution, I would appreciate >>> your considering my company for it. >> >> Sure, I will be doing a lot of research on that for sure. >> ___ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos >> > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Benefits of VMWare 2.0 over 1.x?
MHR wrote: > I've been running VMWare Server 1.x for some time now, currently on > 1.08, and I've been pretty satisfied with it. > > I was wondering if any of you fellow VMWare users are seeing any > significant benefit to moving to the 2.0 release. I haven't tried doing anything new or different - the main thing is that you don't have to rebuild the kernel module on every kernel update - and I think the clock may be more stable for the VM guests. > I'm running CentOS 5.4 with Linux 2.6.18-164.el5 #1 SMP x86_64 on an > AMD 7750 64x2 with 4GB of RAM If you are going to change anything, you might want to consider installing ESXi natively on the hardware and run even your main Centos host as a VM under it. I have some setups where most of the work is done on the Centos host which also exports it's home directory via NFS and one or more guests map the same home directory for some specialized things. I haven't decided if ESXi would be a win for that setup or not. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] printers... (HP)
m.r...@5-cent.us wrote: > I can't remember if I asked here - I've got more than one printer, HP, > laser, builtin JetDirect. When I go there in a browser, I see the basic > thing, but I can't get to the networking page. It says that it wants a > password, no user name. It doesn't accept anything. Last week, a co-worker > reset one to factory defaults... and it comes up with the same thing. > Looking around on the Web, I don't see anything about a default password. > > Clues for the poor? > > mark > You might find something at http://articles.techrepublic.com.com/5100-10878_11-1054861.html or even at http://www.irongeek.com/i.php?page=security/networkprinterhacking No guarantee! I use Brother. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] combining iptables parameters
Message-ID: On: Wed, 28 Oct 2009 20:32:14 +0100, Marcus Moeller wrote: > Hi all, > > is there a way to combine iptables parameters like: iptables > -A OUTPUT -p UDP & -p TCP -d $IP1 & -d $IP2 ? man iptables ... -p, --protocol [!] protocol The protocol of the rule or of the packet to check. The specified protocol can be one of tcp, udp, icmp, or all, or it can be a numeric value, representing one of these protocols or a different one. A protocol name from /etc/protocols is also allowed. A "!" argument before the protocol inverts the test. The number zero is equivalent to all. Protocol all will match with all protocols and is taken as default when this option is omitted. -s, --source [!] address[/mask] Source specification. Address can be either a network name, a hostname (please note that specifying any name to be resolved with a remote query such as DNS is a really bad idea), a network IP address (with /mask), or a plain IP address. The mask can be either a network mask or a plain number, specifying the number of 1s at the left side of the network mask. Thus, a mask of 24 is equivalent to 255.255.255.0. A "!" argument before the address specification inverts the sense of the address. The flag --src is an alias for this option. If both TCP and UDP are both to be blocked it may be inferred that no other protocol is desired to connect either, so simply not specifying any protocol will block all protocols including both TCP and UDP. If both are allowed then again there seems little purpose in blocking any others and again by not specifying any protocol you will obtain the desired result. If instead you do want to allow ICMP but nothing else then !icmp will match all protocols other than icmp. If IP1 and IP2 belong to a CIDR group and no intervening addresses are permitted access as well then a suitable netblock mask will archive the desired results. If not then each IP must be separately specified. So, something like this should work as a block for all protocols to either address: iptables -A OUTPUT -d $IP1 -j DROP iptables -A OUTPUT -d $IP2 -j DROP -- *** E-Mail is NOT a SECURE channel *** James B. Byrnemailto:byrn...@harte-lyne.ca Harte & Lyne Limited http://www.harte-lyne.ca 9 Brockley Drive vox: +1 905 561 1241 Hamilton, Ontario fax: +1 905 561 0757 Canada L8E 3C3 ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] combining iptables parameters
Dear James. >> is there a way to combine iptables parameters like: iptables >> -A OUTPUT -p UDP & -p TCP -d $IP1 & -d $IP2 ? I should have better written something like: -A OUTPUT -p UDP OR -p TCP -d $IP1 OR -d $IP2 as that's what I was looking for. Sorry. Best Regards Marcus ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS 5.4: yum-priorities seems to be MIA
Hi all, I just installed 5.4 on a machine here... All seems running fine, except I decided to put on the extra repositories - and as usual I decided to install yum-priorities (as directed here: http://wiki.centos.org/AdditionalResources/Repositories/RPMForge ) but it seems like this package is nowhere to be found. Does anybody know why this would be? Thanks. Boris. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] resolv.conf rewritten every reboot. How to figure out who and why?
On Thu, Oct 29, 2009 at 9:00 AM, Meenoo Shivdasani wrote: > > > BOOTPROTO=bootp is triggering it. > I'm confused. I just rebooted another machine with 'BOOTPROTO=bootp' in /etc/sysconfig/network-scripts/ifup-eth, it did not rewrite /etc/resolv.conf. Okay, my goof, the one with the problem is actually running fc9. Sorry. I will try sed -i "s/BOOTPROTO=bootp/BOOTPROTO=none/" /etc/sysconfig/network-scripts/ifup-eth mahalo, Dave ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> Is software RAID 10 decent performance? > depends on the workload. committed random writes are greatly accelerated by battery backed write caches on real raid controllers.this greatly speeds up things like transactional databases. if your workload is mostly read, software raid performs quite adequately. with 4 x raid10, you can be doing 4 different reads at once, or two writes simultaneously, as long as you have plenty of pending IO requests. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS 5.4: yum-priorities seems to be MIA
On Thu, Oct 29, 2009 at 1:00 PM, Boris Epstein wrote: > Hi all, > > I just installed 5.4 on a machine here... All seems running fine, > except I decided to put on the extra repositories - and as usual I > decided to install yum-priorities (as directed here: > http://wiki.centos.org/AdditionalResources/Repositories/RPMForge ) but > it seems like this package is nowhere to be found. Does anybody know > why this would be? It is a known issue: http://bugs.centos.org/view.php?id=3923 Akemi ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Not able to FTP since 5.4
> From: Barry Brimer [mailto:li...@brimer.org] > Sent: Wednesday, October 28, 2009 11:33 PM > To: CentOS mailing list > Subject: Re: [CentOS] Not able to FTP since 5.4 > > > I do not use ftp much lately, but my bro did and noticed we > cannot ftp > > to the server since the upgrade. > > > > Using VSFTP. > > Tried rebooting but nothing. > > > > Looks like it goes through the whole process and then 'bam'. > > Could not find an error log that listed the error anywhere. > > > > Worked great before 5.4 update, now it does not work at all. > Apparetnly this is a bughave to set port connect to 20 = yes...worked fine after that. Unknown if this will be fixed or they fixed something that makes this now necessary. On bugzilla I found out.. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> Is software RAID 10 decent performance? Given that you are just starting out, go with SW raid10. When your usage grows, plan to move to hardware raid or a hosted solution. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> $700, eesh. You can get some nice Areca cards for much less than that. What is the cheapest SATA hardware raid card I can get at newegg.com? Seems like most turn out not to be true hardware raid that I have found and will not run on CentOS 4.8 without a great deal of grief. Matt ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Matt: Everyone I know recommends Areca cards. Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime > -Original Message- > From: centos-boun...@centos.org > [mailto:centos-boun...@centos.org] On Behalf Of Matt > Sent: Thursday, October 29, 2009 3:44 PM > To: CentOS mailing list > Subject: Re: [CentOS] Infrastructure HELP! > > > $700, eesh. You can get some nice Areca cards for much > less than that. > > What is the cheapest SATA hardware raid card I can get at newegg.com? > Seems like most turn out not to be true hardware raid that I have > found and will not run on CentOS 4.8 without a great deal of grief. > > Matt > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
On Thu, 29 Oct 2009, Matt wrote: > What is the cheapest SATA hardware raid card I can get at newegg.com? > Seems like most turn out not to be true hardware raid that I have > found and will not run on CentOS 4.8 without a great deal of grief. Not a direct answer to your question, but be careful of SATA drives. As I understand it SAS drives (Serially Attached SCSI) are designed to handle server duty (multiple processes stepping the head 24/7), whereas SATA drives (Serially attached ATA) are not. SATA drives if used in servers will fail prematurely. They're great/cheap for home use though. I've also heard that SATA-2 drives are more like SAS or SCSI drives in this respect, as in they're designed for server duty. I probably heard the above on this very list in the past. Can someone confirm or deny? Making me look like a fool is ok, I'm used to it. ;-) -- Curt Mills, WE7U hacker at fluke dot com Senior Methods Engineer/SysAdmin "Lotto: A tax on people who are bad at math." -- unknown "Windows: Microsoft's tax on computer illiterates." -- WE7U "The world DOES revolve around me: I picked the coordinate system!" Please be advised that this email may contain confidential information. If you are not the intended recipient, please do not read, copy or re-transmit this email. If you have received this email in error, please notify us by email by replying to the sender and by telephone (call us collect at +1 202-828-0850) and delete this message and any attachments. Thank you in advance for your cooperation and assistance. In addition, Danaher and its subsidiaries disclaim that the content of this email constitutes an offer to enter into, or the acceptance of, any contract or agreement or any amendment thereto; provided that the foregoing disclaimer does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment to this email. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] printers... (HP)
someone wrote: > m.r...@5-cent.us wrote: >> I can't remember if I asked here - I've got more than one printer, HP, >> laser, builtin JetDirect. When I go there in a browser, I see the basic >> thing, but I can't get to the networking page. It says that it wants a >> password, no user name. It doesn't accept anything. Last week, a >> co-worker >> reset one to factory defaults... and it comes up with the same thing. >> Looking around on the Web, I don't see anything about a default >> password. >> >> Clues for the poor? >> > You might find something at > http://articles.techrepublic.com.com/5100-10878_11-1054861.html > > or even at > http://www.irongeek.com/i.php?page=security/networkprinterhacking > Could get there - security blocks it. However, one of the other links folks mentioned worked: I did a cold reset (power off, press and hold the "go" button, power on, till you see the "cold reset" come on the panel. Thanks, all! mark > No guarantee! I use Brother. > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS 5.4: yum-priorities seems to be MIA
On Thu, Oct 29, 2009 at 4:07 PM, Akemi Yagi wrote: > On Thu, Oct 29, 2009 at 1:00 PM, Boris Epstein wrote: >> Hi all, >> >> I just installed 5.4 on a machine here... All seems running fine, >> except I decided to put on the extra repositories - and as usual I >> decided to install yum-priorities (as directed here: >> http://wiki.centos.org/AdditionalResources/Repositories/RPMForge ) but >> it seems like this package is nowhere to be found. Does anybody know >> why this would be? > > It is a known issue: > > http://bugs.centos.org/view.php?id=3923 > > Akemi > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > Thanks... Hopefully that will be sorted out soon then. Boris. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Curt Mills wrote: > On Thu, 29 Oct 2009, Matt wrote: > > >> What is the cheapest SATA hardware raid card I can get at newegg.com? >> Seems like most turn out not to be true hardware raid that I have >> found and will not run on CentOS 4.8 without a great deal of grief. >> > > Not a direct answer to your question, but be careful of SATA drives. > > As I understand it SAS drives (Serially Attached SCSI) are designed > to handle server duty (multiple processes stepping the head 24/7), > whereas SATA drives (Serially attached ATA) are not. SATA drives if > used in servers will fail prematurely. They're great/cheap for home > use though. > > I've also heard that SATA-2 drives are more like SAS or SCSI drives > in this respect, as in they're designed for server duty. > > I probably heard the above on this very list in the past. > > Can someone confirm or deny? Making me look like a fool is ok, I'm > used to it. ;-) > > We use SATA in a number of Production Systems. If its server class hardware I haven't seen issues. So I wouldn't treat this as a big concern. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] combining iptables parameters
On Wednesday 28 October 2009 16:36, Marcus Moeller wrote: > Dear Ryan. > > >> is there a way to combine iptables parameters like: iptables -A OUTPUT > >> -p UDP & -p TCP -d $IP1 & -d $IP2 ? > > > > Each of those parameters is called a "match", in IPTables-speak. You > > can specify multiple matches in one rule, but all matches are combined > > with an implicit logical AND. There is no way to get a logical OR > > amongst multiple matches in a single rule. If you want OR logic, you > > use multiple rules. > > > > So, your example could not work as single rule, because no single IP > > packet can be both TCP and UDP, and no single IP packet can have > > multiple destination IP addresses. IPTables tries to prevent you from > > creating nonsensical rules like that in most situations. > > > > You would have to specify the required match space across multiple > > rules, maybe something like this: > > > > iptables -A OUTPUT -p UDP -d $IP1-j DROP > > iptables -A OUTPUT -p TCP -d $IP1 -j DROP > > iptables -A OUTPUT -p UDP -d $IP2 -j DROP > > iptables -A OUTPUT -p TCP -d $IP2 -j DROP > > That's what I am doing atm. Thanks for the update. Even simpler; iptables -A OUTPUT -d $IP1 -j DROP iptables -A OUTPUT -d $IP2 -j DROP This will catch everything doesn't matter if its UDP or TCP or ICMP. -- Regards Robert Linux User #296285 http://counter.li.org ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
> Not a direct answer to your question, but be careful of SATA drives. Seagate does make enterprise SATA drives. -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] grub problems
On Wednesday 28 October 2009 15:47, m.r...@5-cent.us wrote: > This is irritating: I've got a server I just upgraded to 5.4, then > rebooted, only to discover that it just *sits* there at the grub boot > menu. I looked at grub.conf, and uncommented hiddenmenu (which should have > been done long ago). > > It *still* sits there when I reboot. Any clues, folks? > > mark > > grub.conf: > #boot=/dev/sda > default=0 > timeout=5 > splashimage=(hd0,0)/grub/splash.xpm.gz > hiddenmenu > title CentOS (2.6.18-164.2.1.el5) > root (hd0,0) > kernel /vmlinuz-2.6.18-164.2.1.el5 ro root=LABEL=/ rhgb > initrd /initrd-2.6.18-164.2.1.el5.img > title CentOS (2.6.18-164.el5) > root (hd0,0) > kernel /vmlinuz-2.6.18-164.el5 ro root=LABEL=/ rhgb > initrd /initrd-2.6.18-164.el5.img > > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos Looks OK. Can you boot this system at all? Have you tried to boot yhe old kernel, maybe there is an issue with the new one? If you cannot boot ayt all try a live CD and check the log files for a clue. -- Regards Robert Linux User #296285 http://counter.li.org ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
On Fri, 2009-10-30 at 09:57 +1300, Clint Dilks wrote: > > Curt Mills wrote: > > On Thu, 29 Oct 2009, Matt wrote: > > > > > >> What is the cheapest SATA hardware raid card I can get at newegg.com? > >> Seems like most turn out not to be true hardware raid that I have > >> found and will not run on CentOS 4.8 without a great deal of grief. > >> > > > > Not a direct answer to your question, but be careful of SATA drives. > > > > As I understand it SAS drives (Serially Attached SCSI) are designed > > to handle server duty (multiple processes stepping the head 24/7), > > whereas SATA drives (Serially attached ATA) are not. SATA drives if > > used in servers will fail prematurely. They're great/cheap for home > > use though. > > > > I've also heard that SATA-2 drives are more like SAS or SCSI drives > > in this respect, as in they're designed for server duty. > > > > I probably heard the above on this very list in the past. > > > > Can someone confirm or deny? Making me look like a fool is ok, I'm > > used to it. ;-) > > > > > We use SATA in a number of Production Systems. If its server class > hardware I haven't seen issues. So I wouldn't treat this as a big concern. there actually are server and consumer grade SATA drives and you should be very careful about what you are buying. Then there's the thought that any SATA drive runs the risk of data loss in a RAID-5 or RAID-6 setup... http://www.enterprisestorageforum.com/technology/features/article.php/3839636 Craig -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Am 29.10.2009 um 21:50 schrieb Curt Mills: > On Thu, 29 Oct 2009, Matt wrote: > >> What is the cheapest SATA hardware raid card I can get at newegg.com? >> Seems like most turn out not to be true hardware raid that I have >> found and will not run on CentOS 4.8 without a great deal of grief. > > Not a direct answer to your question, but be careful of SATA drives. > > As I understand it SAS drives (Serially Attached SCSI) are designed > to handle server duty (multiple processes stepping the head 24/7), > whereas SATA drives (Serially attached ATA) are not. SATA drives if > used in servers will fail prematurely. They're great/cheap for home > use though. > > I've also heard that SATA-2 drives are more like SAS or SCSI drives > in this respect, as in they're designed for server duty. > > I probably heard the above on this very list in the past. > > Can someone confirm or deny? Making me look like a fool is ok, I'm > used to it. ;-) There are server-grade SATA2 drives. They're usually a bit louder but are designed to run 24/7. They can also deal better with vibrations. Desktop-drives are designed to run for maybe 10hours a day - but they can achieve more spin-up/spin-down cycles than server-grade SATA-drives. SAS drives deliver more IOP/s. You waste a lot of SATA-capacity to achieve the same amount of IOP/s. Really, for the stuff he wants to do, he does not need a lot of I/O - IMO. Without knowing what kind of social networking site he has in mind, I'd say that the traffic of 5000 users (of course, not 5000 simultaneous users) can easily be handled by a single server with enough RAM. Rainer ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
On Wednesday 28 October 2009 16:44, Marcus Moeller wrote: > does it work to define iptables rules with a fqdn as destination > instead of an IP address? Or is it useful to resolve the name first > using e.g. nslookup, writing the result to a variable which is then > used within the -d statement? Best bet it to stay with the address. -- Regards Robert Linux User #296285 http://counter.li.org ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] printers... (HP)
m.r...@5-cent.us wrote: > someone wrote: > >> m.r...@5-cent.us wrote: >> >>> I can't remember if I asked here - I've got more than one printer, HP, >>> laser, builtin JetDirect. When I go there in a browser, I see the basic >>> thing, but I can't get to the networking page. It says that it wants a >>> password, no user name. It doesn't accept anything. Last week, a >>> co-worker >>> reset one to factory defaults... and it comes up with the same thing. >>> Looking around on the Web, I don't see anything about a default >>> password. >>> >>> Clues for the poor? >>> >>> >> You might find something at >> http://articles.techrepublic.com.com/5100-10878_11-1054861.html >> >> or even at >> http://www.irongeek.com/i.php?page=security/networkprinterhacking >> >> > Could get there - security blocks it. > > However, one of the other links folks mentioned worked: I did a cold reset > (power off, press and hold the "go" button, power on, till you see the > "cold reset" come on the panel. > > Thanks, all! > >mark > All's well that ends. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Upgrading Centos 5.3 -> 5.4
Thanks for your help Bart. I had to download the device-mapper RPMs and manually upgrade them first. Once I'd done that, the upgrade went fine. Bart Schaefer wrote: The mirrors that are being selected for you must not be fully up to date. I don't know why that would be the case, but from a not-yet-updated (actually still 5.2) system I get: Loading "priorities" plugin Loading "fastestmirror" plugin Loading mirror speeds from cached hostfile * rpmforge: ftp-stud.fht-esslingen.de * base: mirror.facebook.net * updates: mirror.facebook.net * addons: mirror.facebook.net * extras: centos-distro.cavecreek.net 444 packages excluded due to repository priority protections Installed Packages device-mapper.i386 1.02.24-1.el5 installed device-mapper-event.i386 1.02.24-1.el5 installed device-mapper-multipath.i386 0.4.7-17.el5 installed Available Packages device-mapper.i386 1.02.32-1.el5 base device-mapper-event.i386 1.02.32-1.el5 base device-mapper-multipath.i386 0.4.7-30.el5 base ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Craig White wrote: > there actually are server and consumer grade SATA drives and you should > be very careful about what you are buying. > an important consideration is error handling.cheap consumer SATA drives tend to delay error reporting as long as they physically can, doing many retries to try and recover the data themselves without host system(controller) intervention. the RAID controller would rather know about the error itself. enterprise and server grade drives are qualified by the raid vendors to operate correctly with the controller error handling mechanisms. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Infrastructure HELP!
Looks like the price has gone up with the economy starting to recover. I paid $68 per WD RE3 500GB drive on Amazon.com back in June. I would still recommend going with 4 drives in RAID 10 over 2 in RAID 1 or even 3 drives in RAID 5. You will get almost double the performance due to being able to stripe across the drives. RAID 1 is just a mirror so you are only going to see the performance of a single drive. I have experience with software RAID levels 1 and 5 with mdadm. My file server has a RAID 1 of 2 x 160GB Seagate SATA Drives and a RAID 5 of 3 x 1TB Hitachi SATA drives. For the RAID 1 drives hdparm shows approx 68MB/s for each drive and 68MB/s for the array. With the RAID 5 I am seeing 82 MB/s for the drive and 140MB/s for the array. Keep in mind this is an older Pentium D with the drives connected to an Supermicro LSI 1068 SAS controller. My ESXi box has a RAID 10 of 4 x 500GB Western Digital RE3 drives on a PERC 6/i controller. The PERC is a Dell branded LSI controller. Inside the virtual machines I am getting 148MB/s. Unfortunately I can't test this at the ESXi level, but the MB/s will defiantly be higher as the virtual machine has another layer to go through. Ryan On Thu, Oct 29, 2009 at 3:25 PM, ML wrote: > Ryan, > >> If you want performance stick with RAID 10. In general the more drives >> (spindles) the faster the array. The Western Digital RE3 500 GB drives >> are a good deal. You should be able to get 4 of those in the low >> $200s. In RAID 10 this would give you better performance than 2 x 1TB >> in RAID 1. >> > They are like $89.99 a piece on NewEgg. I have a friend that has 1 x > 1TB Seagate Raid level drives he will sell me for $100 each. > > Is software RAID 10 decent performance? > > -Jason > >> Ryan >> >> On Thu, Oct 29, 2009 at 2:36 PM, ML >> wrote: >>> >>> On Oct 29, 2009, at 11:11 AM, Neil Aggarwal wrote: >>> RAID 10 is striping across mirrored drives. So, if you have 4 x 1TB drives, think of it as two separate 1 TB volumes. The system will write half your data to volume A and the other half to volume B. The data in volume A and B do not overlap. Now, each volume is composed of a mirrored set of drives. Anything written to volume A is actually stored on two drives. Anything written to volume B is actually stores on the other two drives. Does this make sense? Let me know if you need any more explanation. >>> >>> No it makes sense. >>> >>> I am contemplating if I really need 4 x 1tb in this system. I mean >>> how >>> much space with some photo's, web pages and MySQL take up if there >>> are >>> 5,000 subscribers to start up? >>> >>> Would 2 x 1TB enterprise drives be enough mirrored? >>> Also, when you move to a hosted solution, I would appreciate your considering my company for it. >>> >>> Sure, I will be doing a lot of research on that for sure. >>> ___ >>> CentOS mailing list >>> CentOS@centos.org >>> http://lists.centos.org/mailman/listinfo/centos >>> >> ___ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables -d fqdn instead of IP
On Thu, Oct 29, 2009 at 17:12, Robert Spangler wrote: > On Wednesday 28 October 2009 16:44, Marcus Moeller wrote: > >> does it work to define iptables rules with a fqdn as destination >> instead of an IP address? Or is it useful to resolve the name first >> using e.g. nslookup, writing the result to a variable which is then >> used within the -d statement? > > Best bet it to stay with the address. No offense, Robert, but I don't think yours is a very helpful statement. When someone asks about alternative web servers, do we just tell them "Best bet is to stay with Apache"? That's just an opaque personal prejudice, and it doesn't give the guy asking the question any new or helpful information. I can definitely think of cases where using FQDNs is a better choice, and I have some examples from my own personal experience. So I don't believe that you can say there is a "best" method, for all situations. You might be ignorant of the applicable use cases, but that doesn't mean they don't exist. Marcus can weigh the pros and cons of both methods, for his particular case, and make an informed choice. -Ryan ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] combining iptables parameters
On Thu, Oct 29, 2009 at 16:57, Robert Spangler wrote: > Even simpler; > > iptables -A OUTPUT -d $IP1 -j DROP > iptables -A OUTPUT -d $IP2 -j DROP > > This will catch everything doesn't matter if its UDP or TCP or ICMP. I think you're missing the point of the original question: It's just an example rule set to illustrate multiple matches. Marcus wanted to know, generally, whether IPTables supports logical ORing matches together. And assuming it is a real-world example: Why would you assume he'd want to block ICMP, too? I allow ICMP in a lot of rule sets that forbid just TCP/UDP traffic, so I can check host uptime and link latency without exposing any listening daemons. My routers use a similar ruleset, too: They need to be able to talk ICMP with anybody on the Internet, but not anything else. -R ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] question about mounting external USB HD
Ivan Arteaga wrote: > John R Pierce wrote: > >> Ivan Arteaga wrote: >> >> >>> Hello, >>> >>> I have a server running CentOS 4.7 and I want to add a new USB-HD for >>> backup some data. I made the file system as ext3 with the command: >>> / >>> mkfs -t ext3 /dev/sdb1/ >>> >>> After that I did add a new Volume Group and Logical Volume >>> (/VolGrup01-LogVol01) /on that file system and mounted it as /backup. It >>> worked fine but when I did reboot the server It wont mount /backup and I >>> got the following error even after the system boots and I try to mount >>> it manually: >>> >>> /mount: special device /dev/VolGrup01/LogVol01 does not exist/ >>> >>> I will appreciate any comment or suggestion about the best way to get >>> through this. >>> >>> >>> >> I would *NOT* use LVM on a removable external drive. >> >> if you did a mkfs /dev/sdb1, you should >> >> # mount /dev/sdb1 /backup >> >> and skip LVM entirely. >> >> >> ___ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos >> >> > John, > > I did remove the LVM and mounted the file system i had in /dev/sdb1 as > /backup, It works now... > > [r...@server]# mount /dev/sdb1 /backup > [r...@mail ~]# df -kh > FilesystemSize Used Avail Use% Mounted on > /dev/sdb1 459G 105M 436G 1% /backup > [r...@server]# > > I cant restart the server now but I will have a maintenance window > tomorrow so I will check after the reboot how it goes... I added it to > /etc/fstab anyway: > > //dev/sdb1 /backup ext3 suid,dev,exec 0 0/ > > Thanks in advance for your suggestion. > > --Ivan. > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > > It worked fine. Thanks again! ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Default ACL question (EXECUTE BIT)
Hello everyone, I'm doing some tests with ACL's and even though I can create a "default" ACL for a directory (that includes "rwx" for the default owner), when I finally create a file wihin that directory the execute bit is chopped off: [...@machine ~]$ mkdir mydir [...@machine ~]$ setfacl -d -m u::rwx,g::-,o::- mydir/ [...@machine ~]$ cd mydir [...@machine mydir]$ touch testFile.txt [...@machine mydir]$ ls -l testFile.txt -rw--- 1 joe joe 0 Oct 29 21:14 testFile.txt I don't think umask is involved here. As far as I know umask isn't involved when dealing with default ACL's. Anyhow, I'm pretty sure this is by design (security-wise). Is there any way to override this behaviour? Thanks, Jorge ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Default ACL question (EXECUTE BIT)
Jorge Fábregas wrote: > I don't think umask is involved here. As far as I know umask isn't > involved > when dealing with default ACL's. Anyhow, I'm pretty sure this is by design > (security-wise). Is there any way to override this behaviour? It's been eons since I played with acls, but I thought you can only view acls via getfacl(or other similar commands) ls -l doesn't do anything to show acls, only unix-style permissions. nate ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
