Re: [CentOS] strange diskspace consuming
Try running, "sync". On Tue, Oct 20, 2015 at 3:11 PM, Götz Reinicke - IT Koordinator wrote: > Hi, > > I do a tgz-backup some maildir-folders with n*1000 off files and a lot > of GB in storage. The backuped maildirs are removed after the tar. > > My assumption was, that the free diskspace should be bigger after that, > but from what I get with df, it looks like I'm loosing space. > > Currently the tgz is saved on the same disk/mountpoint. > > Any hint, why removing the maildirs dont free diskspace as expected? > > It is still an ext3 filesystme > > Regards and thanks for hints . Götz > > -- > Götz Reinicke > IT-Koordinator > > Tel. +49 7141 969 82420 > E-Mail goetz.reini...@filmakademie.de > > Filmakademie Baden-Württemberg GmbH > Akademiehof 10 > 71638 Ludwigsburg > www.filmakademie.de > > Eintragung Amtsgericht Stuttgart HRB 205016 > > Vorsitzender des Aufsichtsrats: Jürgen Walter MdL > Staatssekretär im Ministerium für Wissenschaft, > Forschung und Kunst Baden-Württemberg > > Geschäftsführer: Prof. Thomas Schadt > > > ___ > CentOS mailing list > CentOS@centos.org > https://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] lvm snapshot
LVM is terribly flawed and most imprudent piece of any filesystem that you can have for snapshot'ing! The only snapshot-friendly filesystem or volume manager is ZFS. Whether create a single snapshot or few hundred thousand. It is instant and no data is lost. On Thu, Dec 3, 2015 at 12:09 AM, Axel Glienke wrote: > > Hello > > after a lvm snapshot creation and a reboot are all logical volumes are > missing, only swap is present. > > > lvcreate -L 5000M -s -n centos_h1-root_snap /dev/mapper/centos_h1-root > > lvs > LV VGAttr LSizePool Origin Data% > Meta% Move Log Cpy%Sync Convert > centos_h1-root_snap centos_h1 swi-a-s---4,88g root 0,00 > homecentos_h1 -wi-ao 15,00g > rootcentos_h1 owi-aos--- 20,00g > swapcentos_h1 -wi-ao7,62g > var centos_h1 -wi-ao 1000,00g > > > after reboot no complete start, dracut console is running: > >dracut:# > ls /dev/centos_h1/ > swap > > dmsetup info shows swap only, too > > I can reproduce this on 2 different Servers and a Virtual Machine have > same behavior. > > What do i wrong? > > > > Greetings > > Axel > > ___ > CentOS mailing list > CentOS@centos.org > https://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Just need to vent
> Seriously, who is in charge with the UI design in gnome? > > Whoever it is needs to be fired. > > /rant > Most of them were already fired or left cause of under-funding. Like other OSS projects, they are severely understaffed. ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Just need to vent
Alice, your rant is just whining. It's open source. If you don't like it, modify it. You cannot compare something which you basically get for free to what is funded by for-profit companies. On Mon, Jan 25, 2016 at 2:32 PM, Alice Wonder wrote: > On 01/25/2016 01:28 AM, Kahlil Hodgson wrote: > >> I personally love Gnome3 on Fedora. It took me about a week to adjust my >> mindset though -- I did that over a Xmas break. >> >> It did help that I read the release notes first (so I was not surprised at >> the major change) and went through the tutorial the developers provided. >> >> An interesting exercise re-examining and critiquing old workflows and >> exploring alternatives. It works really well on the smallish laptop that >> I >> use while commuting and which I plug into a couple of monitors when I get >> to work. Its great the way it frees up screen real estate and encourages >> me to focus on "what I am doing" rather than distracting me with "things I >> might want to do". >> >> Reading the release notes before installing an OS is a really good idea. >> Fedora and RedHat do a really good job with their release notes. >> ___ >> CentOS mailing list >> CentOS@centos.org >> https://lists.centos.org/mailman/listinfo/centos >> >> > Some people adapt to new workflows easily, others do not. > > But regardless of the workflow - back to my original point - it's pretty > damn stupid that choosing a font for a text editor not only includes all > fonts regardless of variable width or monospace, but doesn't identify which > fonts in the selection are monospace. > > How can something like that be missed by their QA / UI testing? > > ___ > CentOS mailing list > CentOS@centos.org > https://lists.centos.org/mailman/listinfo/centos > ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
[CentOS] qemu-kvm high cpu usage with idle windows guest
Windows 2k8 R2 guest with updated virtio drivers is idle inside but on the host qemu-kvm process uses 7-15% cpu. Things that have been tried without any significant success - removed tablet device - manually set cpu topology for cores per socket Is it just me or the qemu-kvm has little tolerance with windows? ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] [CentOS-virt] qemu-kvm high cpu usage with idle windows guest
On Sun, Apr 7, 2013 at 8:32 PM, Zoltan Frombach wrote: > Check out these threads they may help you solve your issue with Windows VMs: Well I did. Upstream provides rpms till version 0.12.1.x. Even the RHEV srpm (the one I am using) has the same version. And I don't want to compile both libvirt and qemu-kvm from source since it is the production machine. > http://forum.proxmox.com/threads/5770-Windows-guest-high-context-switch-rate-when-idle > > http://serverfault.com/questions/146922/why-is-idle-windows-vm-using-so-much-cpu > > On 4/7/2013 5:18 PM, Banyan He wrote: >> Maybe you can print out what it is doing, then try to figure out if they >> are doing good. My linux works good, never tried windows. >> Did. Nothing abnormal. A quick perf shows this: # .. .. # 4.96% qemu-kvm [kvm] [k] kvm_arch_vcpu_ioctl_run 4.83% qemu-kvm [kvm_intel] [k] vmx_vcpu_run 4.47% qemu-kvm [kernel.kallsyms] [k] find_busiest_group 3.23% qemu-kvm [kernel.kallsyms] [k] _spin_lock 3.10% qemu-kvm [kernel.kallsyms] [k] native_write_msr_safe 1.24% qemu-kvm [kernel.kallsyms] [k] _spin_lock_irq 1.22% qemu-kvm [kvm] [k] paging64_walk_addr 1.21% qemu-kvm [kvm] [k] x86_decode_insn 1.21% qemu-kvm [kvm_intel] [k] vmcs_writel >> Banyan He ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Cert RHCSA
Unless you really wanna spend money on it, I would say that certification is just a piece of paper. Ah discard that, in this case, it is not even a piece of paper since you don't get one after passing! Have been there, seen the guys who knew more than CEs or SAs. Just get the books, read them, study the docs and put it to use! On Sun, Jun 2, 2013 at 8:45 PM, Justin Edmands wrote: > > > > On Jun 2, 2013, at 11:19 AM, Eliezer Croitoru wrote: > >> On 6/2/2013 3:58 AM, Fidel Dominguez wrote: >>> Hello everyone >>> >>> I'm already on the training to get the certification on RHCSA. and I >>> wondered if someone has a guide or something else from where to study. >>> >>> Thks >>> Fidel Dominguez >>> ___ >>> CentOS mailing list >>> CentOS@centos.org >>> http://lists.centos.org/mailman/listinfo/centos >> There are couple books you can buy and read. >> Also RH has lots of docs. >> I have seen nice video tutorials and training which you can buy. >> >> What are looking for? >> >> Eliezer >> ___ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos > > I'd say the Michael Jang RHCSA/RHCE book. > > http://www.amazon.com/gp/aw/d/0071765654/ref=mp_s_a_1_1?qid=1370187824&sr=1-1&pi=SL75 > ___ > 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] file system replication
Most people forget that DRDB is a 'clustered' storage that creates a new block device on top of existing block devices, over the network. There are issues like SPLIT BRAIN and a possible data inconsistency, should there be multiple network timeouts, congestions (since it is all network based) etc. This may not be the best solution for your problem! ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Best way to duplicate a live Centos 5 server?
The best and traditional way that has been there for decades is an rsync and then reinstallation of boot-loader. It works always if you know how it's done. If you need detailed instructions, I can send you that! On Tue, Jun 19, 2012 at 10:44 AM, Les Mikesell wrote: > On Tue, Jun 19, 2012 at 12:03 PM, Cal Sawyer wrote: > > > > You're right - documentation is pretty dire. Guess i'm not alone in > > hating doing it. > > Yes, I really, really wish the stuff they are doing was documented, > somewhere, anywhere. Not just how to use the program itself which is > supposed to 'just work' unattended once you set it up, but the black > magic of how they detect and reproduce all of the hardware, lvm, raid, > filesystem, etc. across different distributions and versions. > > > USB backup is broken due to the order in which path variables get set - > > sure is lot of fun trawling through the scripts to find out what gets > > set when. Hope the ReaR maintainers are interested in this and haven't > > gotten themselves mired in tape > > archive integration - i would have thought USB backup was the winner in > > terms of getting broad acceptance as a bare-metal DR solution. > > USB is sort of 'hands-on' for something that should be unattended, and > adds a lot of unpredictable messiness in drive detection, boot order, > etc.All you really have to do is export some NFS space and point > ReaR to it. At least that is the easy way to get started. If you > have another Linux box, just plug your USB drive in there and access > it over NFS... problem solved. > > Clonezilla-live plays in this space too, but it doesn't do raid or > multiple disks at once, and you have to shut down to take the image. > > My 'ideal' system would be to have ReaR generate a directory of what > will be on the boot iso leaving that somewhere on the host without > actually making the image. Then use backuppc to back up the whole > host and its normal duplicate-pooling mechanism will keep the extra > copies of the tools from taking extra space. Then when/if you need a > bare-metal restore, you would first grab the directory of the iso > contents, burn a boot CD, let that reconstruct the filesystem, then > tell backuppc to restore to it. That way would be completely > automatic and always be up to date, with the advantage of backuppc's > efficient storage and easy online access to individual files and > directories. If you don't mind wasting a small amount of space for > the isos, I think that approach would already work if you tell ReaR to > just make the boot image and to wait for an external program to do the > restore after the filesystem has been rebuilt. > > > However, when it works - wow. Just restored an HP dl360 w/HWRAID to a > > Presario desktop machine and it lives! No network, but that's small > > beans compared to the larger win. > > Yes, I've even modified the filesystem layout file to go from a > software RAID to a non-RAID, and to change partition sizes during the > restores. If it was documented, that capability by itself would be > fantastic. > > -- > Les Mikesell > lesmikes...@gmail.com > ___ > 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] Hacking Issue
On Mon, Sep 26, 2011 at 10:10 AM, wrote: > Theo Band wrote: > > On 09/26/2011 01:02 PM, Jennifer Botten wrote: > >> > >> I am having an issue with someone accessing our server via a SIP/VOIP > >> connection. I have changed my iptables rules to drop all UDP traffic > >> from and too this IP address, but this traffic seems to still run > >> through my server. These are the iptables rules that I current have on > >> the server. > >> > >> -A INPUT -i eth0 -s 209.61.231.42 -p udp -j DROP > >> > >> -A INPUT -i eth0 -d 209.61.231.42 -p udp -j DROP > >> > >> > > If your SIP server needs to be accessed from any IP address, consider to > > use fail2ban. Easy to setup and it will block access to your SIP server > > after so many false attempts. > > I started using fail2ban to prevent the logs (Asterisk) from cluttering > > failed logons. > > Let me chime in: *yes* to fail2ban. We use it here at work, and it works, > and is very good. Not too hard to configure for basic usage, either, but > very extensible. > > mark > > I use CSF and LFD. Like every other firewall, the backend is always iptables but CSF is so powerful and has a lot of irreplaceable power and functionality. It can be a bit of difficult to configure it for the first time but when you do, you will never look back ;) -Micky. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] Files being modified in /bin/
For the binary experts. I have a situation here. Something hideously but continuously is modifying the /bin/ executables as common as coreutils and net-tools. I can verify that from md5sum. First thing I checked was 'ls' and it has a checksum mismatch. So I removed it and reinstalled it. Then I moved the file somewhere else to cross bisect it. I did a hexdump on original ls file and the modified file, and there was some 700 lines of hex code additional in the modified file. Then I set a cron to check and do md5sum on all system files and after half an hour, I go a report back. Files modified. This time when checked the hex dump of newly and earlier modified files, they were the same. Exact same! Because rpm and rpmverify also seemed to have been modified so I cannot trust 'rpm -V' package verification. Already did lsof and process tracing but to no avail. Does anyone have any idea how to find that culprit? -Micky. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Files being modified in /bin/
So apparently prelink was running. I disabled it in /etc/sysconfig/prelink and ran 'prelink -ua' to undo the linking. I just stumbled upon a document (attached) describing how Linux used to have a.out <http://en.wikipedia.org/wiki/A.out> and now the ELF. Though I never knew that prelink actually modifies the files and thought of it as a cache library or something. Literally modifies!! So, I assume the problem is solved as ls seems to have reverted back but if not then it may be an LKM kit :| On Mon, Sep 26, 2011 at 6:29 AM, Rob Kampen wrote: > Jeremy Sanders wrote: > >> Micky L Martin wrote: >> >> >> >>> Because rpm and rpmverify also seemed to have been modified so I cannot >>> trust 'rpm -V' package verification. >>> >>> Already did lsof and process tracing but to no avail. Does anyone have >>> any >>> idea how to find that culprit? >>> >>> >> >> Are you sure it's not prelink that's modifying the files? You can google >> how to disable this. >> >> > Any comments or thoughts from the list as to the benefit of prelink? > does the system performance change if this is disabled? > It causes issues with aide also. > > Boot from a CD to check the checksums or run rpm if you want a clean >> environment. >> >> Jeremy >> >> >> __**_ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/**mailman/listinfo/centos<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] Problems with Intel Ethernet and module e1000e
In our case, it was the driver. Google e1000e-1.6.2.tar.gz On Mon, Sep 26, 2011 at 6:56 AM, wrote: > Patrick Lists wrote: > > On 09/23/2011 12:54 PM, Volker Poplawski wrote: > >> Hi all, > >> > >> I'm facing a serious problem with the e100e kernel module for Intel > >> 82574L gigabit nics on Centos 6. > > > > I have also had problems with an Intel Gbit nic in a Dell Optiplex 760 > > using the e1000 module. If it would come up it would come up at 100Mbit > > speed. Only unplugging the LAN cable en reinserting it in the switch > > would sometimes result in the proper 1Gb speed. After replacing the 3Com > > Gbit switch with a DLink Gbit switch things started to work ok. > > > > As suggested, you could try out the e1000 kernel module form elrepo (if > > it works, good for you) but you may want to look into the switch too. > > Hmmm, we had a problem like that, but in our case, it was the NIC. If > someone's already asked this - I didn't read the list over the weekend - > ignore the request, but if not, while it's not working, what's the o/p of > ethtool eth > >mark > > ___ > 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] Files being modified in /bin/
No Jeremy, reformatting is nonsensical, like doing anything without finding cause of the problem is! You have to check out prelink if you still don't know about it, it can be something amazing or ridiculous. In my case, all evidence points to prelink! To the guys using prelink and having experience with it. So I did a further study and found out that it possesses some issues. First it doesn't randomly address the data making applications prone to perl security attacks. Secondly, it way of keeping track of address maps is awkward which becomes a few weeks older till it gets updated. Thirdly, its 'a very old styled' application. It was written back in 90's when computers were slow to make them fast. But with today's extraneous processing age, its effects are long vanishing. Lastly, I see a lot of people remove it in post installation process. Many claim it sucks as it creates more problems than what it is supposed to do. As I occasionally do a minimal install so I am not sure how it got installed on this very box. Seems like 'yum update' or the kickstart did it. But anyhow I disabled it already and I am gonna benchmark the system for performance; needless to day, it will be removed from my desk. It did pop like a jack in a box :P! On Mon, Sep 26, 2011 at 7:11 AM, wrote: > Jeremy Sanders wrote: > > Micky L Martin wrote: > > > >> Because rpm and rpmverify also seemed to have been modified so I cannot > >> trust 'rpm -V' package verification. > >> > >> Already did lsof and process tracing but to no avail. Does anyone have > >> any idea how to find that culprit? > > > > Are you sure it's not prelink that's modifying the files? You can google > > how to disable this. > > > > Boot from a CD to check the checksums or run rpm if you want a clean > > environment. > > Don't really know about prelink, but I strongly agree with the last > suggestion: boot from a CD, or USB key, or something *other* than your > hard drive - your comments strongly suggest that you've been infected. You > *do* have backups of your configuration and data (and home directories, > etc)? If so, you might want to do a reinstall without formatting... and > then, and only then, rerun grub-install. > > mark > > ___ > 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] Hacking Issue
I would use an '-I' instead of '-A' if its a case of blocking an intruder. You can use tcpdump and 'ss -l' as well. Check out the application logs, try to see what's the intruder is up to! On Mon, Sep 26, 2011 at 7:14 AM, Keith Roberts wrote: > On Mon, 26 Sep 2011, Jennifer Botten wrote: > > > To: centos@centos.org > > From: Jennifer Botten > > Subject: [CentOS] Hacking Issue > > > > Hi, > > > > > > > > I am having an issue with someone accessing our server via a SIP/VOIP > > connection. I have changed my iptables rules to drop all UDP traffic from > > and too this IP address, but this traffic seems to still run through my > > server. These are the iptables rules that I current have on the server. > > > > -A INPUT -i eth0 -s 209.61.231.42 -p udp -j DROP > > > > -A INPUT -i eth0 -d 209.61.231.42 -p udp -j DROP > > You might find it helps to analyse this traffic with a > network analyser, like Wireshark. That would allow you to > see in almost real time what is happening on the line. > > Kind Regards, > > Keith Roberts > > - > Websites: > http://www.karsites.net > http://www.php-debuggers.net > http://www.raised-from-the-dead.org.uk > > All email addresses are challenge-response protected with > TMDA [http://tmda.net] > - > ___ > 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
