Re: [CentOS] Detect file change
If you know C, you can write a simple program using inotify(7). For example, you could write a program to continually monitor the directory and pass in the script plus args as a arg. See: http://www.ibm.com/developerworks/linux/library/l-inotify.html Cheers, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] best parallel / cluster SSH
I've been using clusterit for several years for multiple small clusters. It works well and was easy to install. I believe I got the Fedora source RPM and rebuilt it for CentOS. On Thu, Feb 4, 2010 at 1:50 PM, Gavin Carr wrote: > On Thu, Feb 04, 2010 at 01:31:39PM -0500, Alan McKay wrote: >> > It depends on what you need to do. If you really have enough machines >> > or long-running jobs that a shell loop through them isn't practical, you >> > might want something higher-level like cfengine or puppet, or at least >> > something running under cron to make them independent. >> >> cfengine or puppet (or something else - slackmaster?) are where I want >> to be eventually - but in the immediate term something like this would >> help a lot. e.g bouncing my 4 front-end apache servers on 4 >> different boxes. That sort of thing. >> >> I was actually going to start another "configuration management redux" >> thread as a follow up to a thread I started a few months ago. > > We've tried all the ones in that article you mentioned, and are currently > using classh - http://freshmeat.net/projects/classh - which is pretty nice, > in spite of being labelled alpha. > > I've packaged it for c5 here: > > http://www.openfusion.com.au/mrepo/centos5-i386/RPMS.of/classh-0.092-1.of.el5.noarch.rpm > > if you'd like to give it a whirl. > > Cheers, > Gavin > > ___ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > Cheers, -- Wade ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Firefox 7 on CentOS 5
Anyone have any luck running updated Firefox on CentOS 5? We have CentOS 5 workstations and need to update their Firefox but it dumps core even when using the libstdc++ from Fedora 9. I would like anything stable and secure beyond Firefox 3 but can't find tarballs for anything other than 7.0.1. CentOS 5.5 i386 w/ updates firefox-7.0.1.tar.bz2 libstdc++-4.3.0-8.i386.rpm from Fedora 9 Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Firefox 7 on CentOS 5
Turns out it works find as a user installed in their directory but not as root when installed in /usr/local/firefox. Not sure what's up but I'm good for now. On my CentOS 5 workstation, I see some artifacts when displaying graphical bullets, especially at specific resolutions. This was supposedly fixed in Firefox 4 but I still see them in Firefox 7. Not major but an annoyance. Cheers, -- Wade Hampton On Wed, Oct 5, 2011 at 9:59 AM, Lars Hecking wrote: > Wade Hampton writes: >> Anyone have any luck running updated Firefox >> on CentOS 5? We have CentOS 5 workstations >> and need to update their Firefox but it dumps >> core even when using the libstdc++ from Fedora 9. >> >> I would like anything stable and secure beyond >> Firefox 3 but can't find tarballs for anything >> other than 7.0.1. >> >> CentOS 5.5 i386 w/ updates >> firefox-7.0.1.tar.bz2 >> libstdc++-4.3.0-8.i386.rpm from Fedora 9 > > Yes, this combination works fine here. Except, I'm using it on 5.7 with > the x86_64 version of everything. Make sure to rename the stdc++ library > to libstdc++.so.6 and drop it into the top firefox installation directory > where the binary and the bulk of the .so libs live. > > > > ___ > 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] Word Perfect [Was: Novell sale news?]
Bit of info on WordPerfect and Linux. WP was ported to SCO Xenix by SDC of Utah in the early 90's. Several of us figured out how to get it running on Linux (1.0 or 1.2 series kernels, libc5) using the SCO libs and iBCS2 the old Intel binary compatible interface (I wrote the Linux WP mini-howto about 15 years ago). The SDC version was WP 6 or 7. They later made native Linux versions of WP 7 and 8. There were several serious problems with macros and some stability problems which AFIK were never fixed. In the late 90's, I wrote numerous long documents in WP on my Linux desktop! For long docs it was far better than Word. Novel bought WordPerfect for $885M in 1993-94 and sold it to Corel for $185M in 1996. See: http://en.wikipedia.org/wiki/WordPerfect After Corel purchased WP, they stopped working with SDC (WP/UNIX/Linux was SDC's major contract if I recall correctly). Corel then released WP9 for Linux using Wine and also made a lot of improvements to wine in the process. This version was nearly identical to the Windows version but seemed slower and had stability issues. However, when Microsoft bailed out Corel, Corel stopped shipping both Corel Linux and WP for Linux. This left a big hole in office suites and was a major impediment to Linux on the desktop for YEARS until OpenOffice came out and was stable (last few of years). A strategic move by Microsoft. Some organizations still use WP (governments, medical, legal, etc.). Most have migrated to Word. BTW, I wrote all the papers for my MBA using macros in DEC Runoff, spooled the papers to 9-track tape, and drove 1/2 mile to the other building to use the DG-controlled 20' long LASER printer Now my kids write all their High School & College papers using OpenOffice and print at home in color. Very different skill set and it gets the job done (less time learning the tool and more focusing on content). Cheers, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Good network printer/scanner for Centos/Linux
HP 6150C scanner/printer. Works well with hplip and cups. Remote scanning works better on Linux than Windows. -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] OpenOffice 3.3.0 crashing on CentOS 5.5
The last few releases of OpenOffice have gotten very unstable on my desktop which is a CentOS 5.5 i386 system. After multiple crashes when doing embedded simple drawings in OpenOffice writer (circles and lines with connections), I started using the versions from Sun then Oracle. I am now using version 3.3: openooffice.org3-3.3.0-9567. I've lived with this but with version 3.x it has gotten really bad. Display info: depth 16-bit, Intel Boradwater-G Chipset 965Q Crashes I am experiencing now: Writer sometimes using 100% CPU and X is unresponsive. (I have to ssh to the workstation and kill Writer to recover.) Random crashes in Writer Writer crashes when moving circles or lines in simple drawings (mainly circles, lines, clipart, etc.) (insert -> object -> OLE object -> OO drawing)... crashes in Calc when adding text at the end of a line of text in a cell. Could this be a problem with some setting or with Java? Any help would be most appreciated, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] OpenOffice 3.3.0 crashing on CentOS 5.5
On Tue, Jun 28, 2011 at 11:17 AM, Ljubomir Ljubojevic wrote:
> Wade Hampton wrote:
>> The last few releases of OpenOffice have gotten very
>> unstable on my desktop which is a CentOS 5.5 i386 system.
>> After multiple crashes when doing embedded simple drawings
>> in OpenOffice writer (circles and lines with connections), I
>> started using the versions from Sun then Oracle. I am
>> now using version 3.3: openooffice.org3-3.3.0-9567. I've
>> lived with this but with version 3.x it has gotten really bad.
>>
>> Display info: depth 16-bit, Intel Boradwater-G Chipset 965Q
>>
>> Crashes I am experiencing now:
>>
>> Writer sometimes using 100% CPU and X is unresponsive.
>> (I have to ssh to the workstation and kill Writer to recover.)
>>
>> Random crashes in Writer
>>
>> Writer crashes when moving circles or lines in
>> simple drawings (mainly circles, lines, clipart, etc.)
>> (insert -> object -> OLE object -> OO drawing)...
>>
>> crashes in Calc when adding text at the end of a
>> line of text in a cell.
>>
>> Could this be a problem with some setting or with Java?
>>
>> Any help would be most appreciated,
>> --
>> Wade Hampton
>> ___
>> CentOS mailing list
>> CentOS@centos.org
>> http://lists.centos.org/mailman/listinfo/centos
>>
>>
> Are there any crash reports?
>
> Run OOO from terminal so you can capture errors and problems.
I ran OOO from the command line: openoffice.org3 -calc
Result was a core dump after entering data on the spreadsheet
for a few minutes.
soffice: line 120: 3882 Segmentation fault
(core dumped) "$sd_prog/$sd_binary" "$@"
gdb soffice.bin core.3882
#0 0x004922aa in ?? () from /opt/openoffice.org3/program/../basis-link/
program/../ure-link/lib/libuno_cppuhelpergcc3.so.3
#1 0x9739df14 in ?? ()
#2 0x000 in ?? ()
This is seems to be from openoffice.org-ure-1.7.0-9567:
/opt/openoffice.org/ure/lib/libuno_cppuhelpergcc3.0.so.3
I found a previous core (29213) file with a similar problem:
#0 0x00472309 in cppu::OWeakObject::acquire() ()
from /opt/openoffice.org3/program/../basis-link/program/../
ure-link/lib/libuno_cppuhelpergcc3.so.3
And another (21260):
#0 0x00275e44 in cppu::WeakComponentImplHelperBase::acquire() ()
from ...{same}
Thanks,
--
Wade Hampton
>
> Ljubomir
> ___
> 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] OpenOffice 3.3.0 crashing on CentOS 5.5
On Tue, Jun 28, 2011 at 1:11 PM, Ljubomir Ljubojevic wrote: > Wade Hampton wrote: >> This is seems to be from openoffice.org-ure-1.7.0-9567: >> /opt/openoffice.org/ure/lib/libuno_cppuhelpergcc3.0.so.3 >> >>> Ljubomir > > I think I know what is going on. Please check if the old version > openoffice.org-ure-3.1.1-19.5.el5_5.6.i386.rpm is "also" installed via > yum since older version is marked for example 3.1.1 in CentOS/RHEL > repository but effectively newer is marked as 1.7.0 in downloaded rpm's. > > I created yum based repository with virtual rpm files that pull all > needed files, and I had to uninstall full OOO so I can remove > openoffice.org-ure-3.1.1 rpm. On this and several other co-worker's computers, I had to remove ALL older OOO files to get it to install. The version is the one from the 3.3 tarball: 000330_m20_native_packed-1_en-use.9567 openoffice.org-ure-1.7.0-9567.i586.rpm > > If that is not it, consider sending me your example hat crashes via > direct e-mail so I can test on my own OOO 3.3 Not sure if I can (corporate computer). I just created a sample spreadsheet with text and numbers and kept typing in text and numbers. Nothing fancy. It crashes after a short time. I have several extensions. Could that be the problem: arrows 3D and some Clipart English spelling and hyphenation French and Spanish dictionary OOOP accessories OOOP templates postgresql-sdbc Professional Template Pack Python Calculator Cheers, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] OpenOffice 3.3.0 crashing on CentOS 5.5
On Tue, Jun 28, 2011 at 1:40 PM, wrote:
> Wade Hampton wrote:
>> On Tue, Jun 28, 2011 at 1:11 PM, Ljubomir Ljubojevic
>> wrote:
>>> Wade Hampton wrote:
>>>> This is seems to be from openoffice.org-ure-1.7.0-9567:
>>>> /opt/openoffice.org/ure/lib/libuno_cppuhelpergcc3.0.so.3
>>>
>>> I think I know what is going on. Please check if the old version
>>> openoffice.org-ure-3.1.1-19.5.el5_5.6.i386.rpm is "also" installed via
>>> yum since older version is marked for example 3.1.1 in CentOS/RHEL
>>> repository but effectively newer is marked as 1.7.0 in downloaded rpm's.
>
>> On this and several other co-worker's computers, I had
>> to remove ALL older OOO files to get it to install. The
>> version is the one from the 3.3 tarball:
>> 000330_m20_native_packed-1_en-use.9567
>> openoffice.org-ure-1.7.0-9567.i586.rpm
>
> *How* did you remove all older OOO files? You are aware that CentOS
Very painfully
> installs OO via yum and rpm (I seem to have the current, which is 3.1). Is
> it possible that you removed tarballs, but not the rpm's?
No, I removed the RPMs and the installed RPMs seem to be OK
(verified by: rpm -qa | grep openoffice | xargs -n1 -t -i rpm -V {} ).
Cheers,
--
Wade
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] OpenOffice 3.3.0 crashing on CentOS 5.5
On Tue, Jun 28, 2011 at 1:55 PM, Ljubomir Ljubojevic wrote: > Wade Hampton wrote: >> Not sure if I can (corporate computer). I just created >> a sample spreadsheet with text and numbers and kept >> typing in text and numbers. Nothing fancy. It crashes >> after a short time. > > I meant something without real data, but that crashes. > >> >> I have several extensions. Could that be the problem: >> arrows 3D and some Clipart >> English spelling and hyphenation >> French and Spanish dictionary >> OOOP accessories >> OOOP templates >> postgresql-sdbc >> Professional Template Pack >> Python Calculator >> > > Then try disabling those extensions one by one. The extension manager hangs when trying to remove. I removed all under the ~/.openoffice.org/3/user/uno_packages and restarted. I'm testing now but I suspect that is the main problem and will post if it was successful or not I still have a problem with selecting regions in drawings and spreadsheets which will cause OO to use 100% CPU -- a separate problem (and quite rare). Thanks, -- Wade ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Temperature sensor
Try the Dallas/Maxim 1-wire system. They have serial port controllers with an RJ11 jack so you can use a phone cable to the sensor. I got one of their temp sensors and a cheap RJ11 jack from Radio Shack and had a remote temp sensor. They use a simple serial protocol and some of the controllers are "smart" like the DS9097U $28 or so for the controller: http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2983 http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2923 For temperature DS28EA00: http://www.maxim-ic.com/quick_view2.cfm/qv_pk/5355 On Fri, Feb 26, 2010 at 2:02 PM, nate wrote: > Dominik Zyla wrote: > >> You have right. While you checking sensors from few machines, you can >> see the trend. Gotta think about changing the way of temperature monitoring >> here. > > Myself I wouldn't rely on internal equipment sensors to try to > extrapolate ambient temperature from their readings. Most equipment > will automatically spin their fans at faster RPMs as the temperature > goes up which can give false indications of ambient temperature. > > I do monitor the temperature of network equipment, but also have > dedicated sensors for ambient readings. Already saved us some pain > once, opened up a new location in London last year and the ambient > temperature at our rack in the data center was 85+ degrees F. The > SLA requires temperature be from 64-78 degrees. Alarms were going off > in Nagios. > > The facility claimed there was no issue, and opened up some more > air vents, which didn't help. They still didn't believe us so they > installed their own sensor in our rack. The next day the temperature > dropped by ~10 degrees, I guess they believed their own sensor.. > > http://portal.aphroland.org/~aphro/rack-temperature.png > > People at my own company were questioning the accuracy of this > sensor(there was only one, I prefer 2 but they are cheap bastards), > but I was able to validate the increased temperature by comparing > the internal temp of the switches and load balancers were > significantly higher than other locations. Though even with the > ambient temperature dropping by 10+ degrees, the temperature of > the gear didn't move nearly as much. > > The crazy part was I checked the temperature probes at my former > company(different/better data center) and the *exhaust* temperature > of the servers was lower than the *input* temperature from this > new data center. Exhaust temperature was around 78-80 degrees, > several degrees below the 85+. > > It seems the facility in London further improved their cooling > in recent weeks as average temperature is down from 78 to about > 70-72 now, and is much more stable, prior to the change we > were frequently spiking above 80 and averaging about 78. > > Also having ambient temperature sensors can be advantageous in > the event you need to convince a facility they are running too > hot(or out of SLA), as a tech guy myself(as you can probably > see already) I am much less inclined to trust the results of > internal equipment sensors than a standalone external sensor > which can be put on the front of the rack. > > nate > > > ___ > 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] Network tuning for working with very old Solaris
G'day. I am trying to get a very old Solaris 5.7 server to ftp data to my updated CentOS 5.4 server but the Solaris box keeps losing networking after sending some data. I can't ping the Solaris box from any of the servers on my network. Could this be related to a much newer (modern) TCP/IP stack in Linux and TCP options? If so, does anyone have any suggestions for how I can tune the Linux server? I am not as concerned about performance, but just to keep the Solaris box from crashing (and no, I can't upgrade the legacy Solaris server). Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Network tuning for working with very old Solaris
Thanks for the feedback. On Linux I have tried setting: rmem_default=32768 rmem_max=65536 tcp_window_scaling=0 tcp_sack=0 tcp_fack=0 I can send data to the Solaris x86 box all day with no problems. It is only when it sends data to my Linux box that it crashes. The NIC is 3Com Etherlink XL PCI. I'm trying the ndd commands in a few minutes after the box reboots yet again On Sat, Mar 27, 2010 at 1:19 PM, JohnS wrote: > > On Sat, 2010-03-27 at 13:02 -0400, Wade Hampton wrote: > >> >> I can't ping the Solaris box >> from any of the servers on my network. > --- > That in it self should tell you to look at the cabling and nic card. I > suspect its a very old nic card like ISA or begining PCI. Just because > it is Sol 5 does not mean TCP/IP will not work. > > ___ > 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] Network tuning for working with very old Solaris
On Sat, Mar 27, 2010 at 1:49 PM, John R Pierce wrote: > JohnS wrote: >> On Sat, 2010-03-27 at 13:02 -0400, Wade Hampton wrote: >> >> >>> I can't ping the Solaris box >>> from any of the servers on my network. >>> >> --- >> That in it self should tell you to look at the cabling and nic card. I >> suspect its a very old nic card like ISA or begining PCI. Just because >> it is Sol 5 does not mean TCP/IP will not work. >> > > SunOS 5.7 is Solaris 7, actually. 'Solaris 10' is really SunOS 5.10. > > Bill Joy wrote the TCP stack thats in all Unix systems, and was copied > for Linux. There's nothing wrong with the TCP in any SunOS version > going as far back as you like. AFIK copied to Windows 2K and later as well > > btw, an older Sun Sparc is more likely to be SBUS and not ISA. If its > PCI, its likely 64bit and maybe 66Mhz. SBUS is a 32bit synchronous > bus thats somewhat slower at clock speed than 32bit 33MHz PCI, however > its capable of higher throughput due to being a more efficient bus protocol. Standard, old PCI. > > me, I'd get on the console (most Sparc's newer than about 10 years old > have a ALOM or RSC or whatever remote console module you can telnet or > ssh to, older ones were almost all serial console, which is typically > connected to a cyclades type console controller you can ssh to), and run > some diags from there (check dmesg for any events, check ifconfig, ndd, > and verify the settings, see if you can ping, etc) Thanks. The Solaris server is on PC hardware and is running CDE. I can log into it even when it can't connect to the network. The ethernet driver is /dev/elx (/dev/elxl0 and /dev/elxl1). The card is a 3Com Etherlink XL PCI card connected to a 3Ccom switch. Trying ndd /dev/elx \? results in "couldn't push module 'elx'. so no idea how to tune it. I can run ndd /dev/ip \? and I get a list of tuneables I've been copying data TO the Solaris box for years without problems (it is a test machine). However, when I try to get data back to my Linux server, the Solaris server seems to lose its networking and I have to reboot it. The last thing that wireshark displays is a bunch of ACK's from my Linux box to the Solaris box (there are multiple connections open). I am about ready to go home as rebooting this Solaris is getting rather old. -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Network tuning for working with very old Solaris
Thanks. I'll try it on Monday when I get back to the machine. On Sat, Mar 27, 2010 at 2:45 PM, John R Pierce wrote: > Wade Hampton wrote: >> Trying ndd /dev/elx \? results in "couldn't push module 'elx'. >> so no idea how to tune it. I can run ndd /dev/ip \? and I get >> a list of tuneables >> > > i'm pretty sure its /dev/elx0 or /dev/elx1 or whatever and not just /dev/elx > > oldest sol box I still have warm is sol9 on a v240 (dual ultrasparc > III), with a bge, and it for sure wants /dev/bge0 > > we never ran much solaris x86 prior to solaris 10, the hardware support > was too sketchy for us. > >> I've been copying data TO the Solaris box for years without >> problems (it is a test machine). However, when I try to get >> data back to my Linux server, the Solaris server seems to >> lose its networking and I have to reboot it. >> > > FWIW, I *never* liked 3Com adapters, I never understood why they were so > popular with the PC crowd. > > > ___ > 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] Building QT on CentOS 5
I am building QT 4.8.6 on CentOS 5 and it is failing on building the webkit module with __sync_add_and_fetch_4 not being defined. My build is for Embedded Linux, which allows me to use the framebuffer. Posts report this error when trying to cross-compile for ARM, but I am building for X86. Several posts also indicate that this is due to the old GCC used by CentOS 5. Does anyone have any suggestions? Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] CentOS 7 installer not seeing SATA disks
I am having trouble with the CentOS 7 installer recognizing SATA hard disks on a Core 2 motherboard. Yesterday I was installing CentOS 7 on a core 2 test computer (rather old machine). The computer had two SATA drives but the installer did not allow me to select them, even to repartition them. During install, I was able to manually run fdisk and sfdisk by accessing a multiscreen using ctrl-alt-F2, so the kernel did see the drives and the /dev entries were present. As a second test, I added a new solid state drive to replace one of the hard disks. The installer saw that drive and allowed me to install on it. Has anyone seen issues like this? Is there a simple workaround? Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Best solution for TRIM on CentOS 5.x
I am planning on using solid state disks with my CentOS 5.x systems. Currently, I am using EXT3 for the file system. >From what I can find, CentOS 5.x does not support TRIM on solid state disks? Is this correct? Should I obtain and build the updated hdparm and use that to trim my drives (for example, via CRON every night at midnight)? Or should I get fstrim (updated util-linux)? Any other suggestions on solid state drives (other than noatime on mounts). Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] SSD support in C5 and C6
I have been following this and have some notes. Can you folks comment on them? I am considering migrating some systems to SSD but have not had time to set up a test system yet to verify it. I found lots of references to TRIM, but it is not included with CentOS 5. However, I found that TRIM is in the newer hdparm which could be build from source, but AFIK is not included with CentOS 5 RPMS. That way, one could trim via a cron job? Could you folks please comment on the below notes that I found from multiple sites online. These are what I was planning on doing for my systems. Notes include: - use file system supporting TRIM (e.g., EXT4 or BTRFS). - update hdparm to get TRIM support on CentOS 5 - align on block erase boundaries for drive, or use 1M boundaries - use native, non LVM partitions - under provision (only use 60-75% of drive, leave unallocated space) - set noatime in /etc/fstab (or relatime w/ newer to keep atime data sane) - move some tmp files to tmpfs (e.g., periodic status files and things that change often) - move /tmp to RAM (per some suggestions) - use secure erase before re-use of drive - make sure drive has the latest firmware - add “elevator=noop” to the kernel boot options or use deadline, can change on a drive-by-drive basis (e.g., if HD + SSD in a system) - reduce swappiness of kernel via /etc/sysctl.conf: vm.swappiness=1 vm.vfs_cache_pressure=50 -- or swap to HD, not SSD - BIOS tuning to set drives to “write back” and using hdparm: hdparm -W1 /dev/sda Any comments? -- Wade Hampton On Fri, Jul 19, 2013 at 10:10 AM, Alexander Arlt wrote: > Am 07/19/2013 03:17 AM, schrieb Lists: > > Main thing is DO NOT EVEN THINK OF USING CONSUMER GRADE SSDs. SSDs are a > > bit like a salt shaker, they have only a certain number of shakes and > > when it runs out of writes, well, the salt shaker is empty. Spend the > > money and get a decent Enterprise SSD. We've been conservatively using > > the (spendy) Intel drives with good results. > > Hm. I'm not sure, if I'd go with that. In my understanding, I'd just buy > something like a Samsung SSD 840 Pro (for not using TLC) and do a > overprovisioning of about 60% of the capacity. With the 512GiB-Variant, > I'd end up with 200GiB netto. By this way, I have no issues with TRIM or > GC (there are always enough empty cells) and wear leveling is also a > non-issue (at least right now...). > > It's a lot cheaper than the "Enterprise Grade SSDs", which are still > basically MLC-SSDs and are also doing just the same as we are. And for > the price of those golden SSDs I get about 7 or 8 of the "Consumer SSD", > so I just swap those out, whenever I feel like it. Or smart tells me to > do so. > > ___ > 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] SSD support in C5 and C6
>From what I have read, TRIM can also be done on demand for older systems or file systems that are not TRIM aware. For CentOS 5.x, a modified hdparm could be used to send the TRIM comamnd to the drive. Anyone have experience with this? -- Wade Hampton On Fri, Jul 19, 2013 at 1:05 PM, John R Pierce wrote: > On 7/19/2013 8:48 AM, Wade Hampton wrote: > > I found lots of references to TRIM, but it is not included > > with CentOS 5. However, I found that TRIM is in the > > newer hdparm which could be build from source, > > but AFIK is not included with CentOS 5 RPMS. That way, > > one could trim via a cron job? > > > trim is done at the file system kernel level.essentially, its a > extra command to the disk telling it this block is complete and the rest > of it 'doesn't matter' so the drive doesn't need to actually store it. > > > On 7/19/2013 7:10 AM, Alexander Arlt wrote: > > Hm. I'm not sure, if I'd go with that. In my understanding, I'd just buy > > something like a Samsung SSD 840 Pro (for not using TLC) and do a > > overprovisioning of about 60% of the capacity. With the 512GiB-Variant, > > I'd end up with 200GiB netto. By this way, I have no issues with TRIM or > > GC (there are always enough empty cells) and wear leveling is also a > > non-issue (at least right now...). > > those drives do NOT have 'supercaps' so they will lose any recently > written data on power failures. This WILL result in corrupted file > systems, much the same as using a RAID controller with write-back cache > that doesn't have a internal RAID battery. > > > > -- > john r pierce 37N 122W > somewhere on the middle of the left coast > > ___ > 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] SSD support in C5 and C6
Thanks for the feedback. Sounds like all this needs to be merged into a wiki? Couple of take-aways: - options will depend on the drive -- cheap drives, be more conservative with options including turning write-cache off -- provisioning depends on how much mfg reserves - better options are available for CentOS 6 - kernel scheduler, swap, and /tmp changes might help for some use cases -- test and determine if they will help (e.g., if your system processes data and creates a lot of files in /tmp for processing, putting /tmp in RAM might help) 1) Determine your use case 2) Determine the type of drive you need and any items specific to the drive (reserved space, TRIM, big caps) 3) Use newer Linux systems (CentOS 6, later UBUNTU, RHEL, Fedora) if you can -- and use EXT4 with trim enabled (if drive supports it) 4) Test 5) Deploy Cheers, -- Wade Hampton On Fri, Jul 19, 2013 at 4:07 PM, Gordon Messmer wrote: > On 07/19/2013 11:21 AM, John R Pierce wrote: > > On 7/19/2013 11:07 AM, Gordon Messmer wrote: > >>> >>- under provision (only use 60-75% of drive, leave unallocated space) > >> >That only applies to some drives, probably not current generation > hardware. > >> > > > it applies to all SSDs. they NEED to do write block remapping, if they > > don't have free space, its much much less efficient.. > > Well, maybe. > > The important factor is how much the manufacturer has over-provisioned > the storage. Performance targeted drives are going to have a large > chunk of storage hidden from the OS in order to support block remapping > functions. Drives that are sold at a lower cost are often going to > provide less reserved storage for that purpose. > > So, my point is that if you're buying good drives, you probably don't > need to leave unpartitioned space because there's already a big chunk of > space that's not even visible to the OS. > > Here are a couple of articles on the topic: > > > http://www.edn.com/design/systems-design/4404566/Understanding-SSD-over-provisioning > http://www.anandtech.com/show/6489/playing-with-op > > Anand's tests indicate that there's not really a difference between > cells reserved by the manufacturer and cells in unpartitioned space on > the drive. If your manufacturer left less space reserved, you can > probably boost performance by reserving space yourself by leaving it > unpartitioned. > > There are diminishing returns, so if the manufacturer did reserve > sufficient space, you won't get much performance benefit from leaving > additional space unallocated. > ___ > 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] SSD support in C5 and C6
On Jul 19, 2013 10:04 PM, "Darr247" wrote: > > On 2013-07-19 1:01 PM, John R Pierce wrote: > > On 7/19/2013 5:51 AM, Darr247 wrote: > >> On 2013-07-19 3:54 AM, Gordon Messmer wrote: > Regardless of your storage, your system should be powered by a > monitored UPS. Verify that it works, and the drive's cache shouldn't > be a major concern. > >> It should also be a 'true sine wave' output when running on battery. > >> Many UPS units output a 'stepped approximation' (typically pulse width > >> modulation), which some computer power supplies may not like. > > virtually all PC and server power supplies now days are 'switchers', and > > could care less what the input wave form looks like. they full wave > > rectify the input voltage to DC, then chop it at 200Khz or so and run it > > through a toroidal transformer to generate the various DC voltages. > > > > > > Heh... go ahead and use stepped approximation UPS's then. > What do I know; I'm just a dumb electrician. I just trust Florida Flicker n Flash - never had outages more than once a day! Sorry could not resist... > ___ > 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] update from local repo mirror
Try running createrepo in your repository directory, the
one with the RPMs in it.
cd {...}/i386
createrepo .
Cheers,
--
Wade Hampton
On Mon, Nov 4, 2013 at 11:05 AM, zGreenfelder wrote:
> so this may be an odd question, but my google fu seems to be failing me.
> I've created a local mirror of centos 5 & 6 with various instructions
> across the net, it seems to be functional, syncs on a regular basis,and is
> generally usable (I've built machines with kickstart/netboot from it and
> they seem generally workable). my difficulty comes up when I try to use
> the local mirror for system updates.
>
> if I use the originally installed repo configs and run yum check-update, I
> get a list of 10-15 RPMs for updates, if I change the repo files, do a yum
> clean all, comment out mirrorlist and uncommenting/chang baseurl=
> http://localmirror/path/to/repo and run a check, I get nothing.
> if I look in localmirror:/path/to/repo and do a find for the version
> numbers I got as available updates from the remote repositories, I find
> appropriate files in seemingly reasonably named directory paths.
>
> my current best guess is that my repodata/repomd.xml file needs some sort
> of update/rebuild via the createrepo script, but I can't seem to find the
> 'right' way to this. does anyone have any pointers?
>
> thanks.
>
> --
> Even the Magic 8 ball has an opinion on email clients: Outlook not so good.
> ___
> 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] Offline YUM repository setup working but yum with --installroot fails
I use a local, off-line repository for CentOS plus updates for my
development network (my setup for many years). By design, the
repository has no physical connection to the Internet. Everything seems
to be setup correctly and works fine for installs/updates. However I am
trying to use the --installroot option for YUM and it is complaining
about not finding a mirror. I removed the lines about mirror from
my CentOS repo files so I am confused.
My CentOS-Base.repo file contains:
[base]
name=CentOS-$releaseserver - Base
baseurl=http://myhost.org/yum/centos/$releasever/os/$basearch/
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-5
[updates]
name=CentOS-$releaseserver - Updates
baseurl=http://myhost.org/yum/centos/$releasever/os/$basearch/
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-5
I have RPMs from the CentOS 5.10 DVDs in a directory indexed by
createrepo. I also have a copy of the current updates (as of Monday).
I can install and update via this configuration.
My web server has the directory tree:
/var/www/html/yum/centos/5 -> /var/www/html/yum/centos/5.10
/var/www/html/yum/centos/5.10/os/i386
/var/www/html/yum/centos/5.10/updates/i386
I am testing some code to make a small, light install. When I
run yum with --installroot, it hangs and eventually times out looking
for a mirror:
{copy centos-release-5-10.el5.centos.i386.rpm to the current dir}
mkdir tmpdir
rpm --root=tmpdir -ivh --nodeps centos-release*rpm
yum --installroot=tmpdir install basesystem filesystem bash kernel passwd
YUM hangs and eventually responds with:
Could not retrieve mirrorlist
http://mirrlist.centos/org/?release=5&arch=i386&repo=os error was
[Errno 4] IOError: http://lists.centos.org/mailman/listinfo/centos
[CentOS] Error starting Virtual Machine Manager: Failed to contact configuration server
I just updated and rebooted a VM host server which runs CentOS 6.5/x86_64. After rebooting, I can't start the virtual-manager due to the error: Error starting Virtual Machine Manager: Failed to contact configuration server; some possible causes are that you need to enable TCP/IP networking for ORBit, or you have stale NFS locks due to a system crash. See http://projects.gnome.org/gconf/ [^] for information. (Details - 1: Failed to get connection to session: Failed to connect to socket /tmp/dbus-Q6...: Connection refused). I have tried: service messagebus restart service libvirtd restart If I manually start dbus, it works and I can run my VMs: dbus-launch --exit-with-session virt-manager Any ideas on how to fix this? -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Error starting Virtual Machine Manager: Failed to contact configuration server
Thanks, however that did not work. I removed /tmp/orbit-root and tried it again. Same problem. For now, I'm just using the dbus-launch trick. Cheers, -- Wade Hampton On Wed, May 14, 2014 at 5:14 PM, Marcelo Roccasalva < marcelo-cen...@irrigacion.gov.ar> wrote: > On Wed, May 14, 2014 at 2:13 PM, Wade Hampton > wrote: > > I just updated and rebooted a VM host server which > > runs CentOS 6.5/x86_64. After rebooting, I can't start the > > virtual-manager due to the error: > > > > Error starting Virtual Machine Manager: Failed to contact configuration > > server; some possible causes are that you need to enable TCP/IP > networking > > for ORBit, or you have stale NFS locks due to a system crash. See > > http://projects.gnome.org/gconf/ [^] for information. (Details - 1: > Failed > > to get connection to session: Failed to connect to socket > /tmp/dbus-Q6...: > > Connection refused). > > I remeber removing some directory from /tmp, maybe > /tmp/orbit- or something like this and reconnecting... Hope > this helps. > > -- > Marcelo > > "¿No será acaso que esta vida moderna está teniendo más de moderna que de > vida?" (Mafalda) > ___ > 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] CentOS 5.8, xen kernel, and nfs4
G'day, I have a workstation running CentOS 5.8 with kernel 2.6.18-308.el5. This workstation needs to mount a NFS4 directory on a server (mysvr in the example below). If the computer is running the XEN kernel (uname -a reports ...2.6.18-308.el5xen), my NFS4 share is mounted but "ls" reports an error. Xen kernel: uname -a ...2.6.18-308.el5xen mount | grep mysvr mysvr:/ on /mnt/mysvr type nfs4 (rw,nodev,hard,intr,addr=1.2.3.4) ls /mnt/mysvr/data ls: /mnt/mysvr/data: Not a directory When I boot to a non-xen kernel, the mount works fine (and has for a few years). uname -a ...2.6.18-308.el5 mount | grep mysvr mysvr:/ on /mnt/mysvr type nfs4 (rw,nodev,hard,intr,addr=1.2.3.4) ls /mnt/mysvr/data myfile.txt ... Any ideas on how I can fix this or should I go back to NFS3 on my server? Thanks, -- Wade Hampton ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
