devel February 2015

devel@lists.centos.org

54 participants
54 discussions

ksplice and CentOS
by Karanbir Singh 25 Sep '25

25 Sep '25

hi guys, lots of people on this list are keen on contributing to the CentOS effort - and one place where we could, as a group of people, make a massive difference is being able to deliver a ksplice upgrade model. A couple of people have started in silo's to try and get something together but I feel it might be a good time to try and see if we can do somethng as a group. Do we have people on the list who find the idea of rebootless kernel upgrades interesting ? Is there some desire to try and make that happen within the CentOS Ecosystem ? - KB

24 47

anaconda dependency calculation?
by Nicholas Byrne 04 Aug '25

04 Aug '25

Hi all, I'm trying to find a tool that calculates all dependencies for a specific package or group of packages for use with CentOS 5 (beta) with the intention of putting together a custom single CD build of CentOS 5. With CentOS 4 i used a python script that used the hdrlist module to find all dependencies which borrowed code from a number of sources. I've looked at yumdownloader and how it checks dependencies (using the --resolve option) but if i pass it the list of packages defined in say the Base and the Core yum groups which generally form an absolute minimal install i end up with about 200 packages. If i do a custom install (using beta 1 disks) and deselect everything i end up with over 300 packages. So i'm guessing anaconda calculates dependencies in a slightly different way, can anyone point me to files/code where i should be looking to find out? Thanks very much in advance for help Nick This e-mail is the property of Quadriga Worldwide Ltd, intended for the addressee only and confidential. Any dissemination, copying or distribution of this message or any attachments is strictly prohibited. If you have received this message in error, please notify us immediately by replying to the message and deleting it from your computer. Messages sent to and from Quadriga may be monitored. Quadriga cannot guarantee any message delivery method is secure or error-free. Information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. We do not accept responsibility for any errors or omissions in this message and/or attachment that arise as a result of transmission. You should carry out your own virus checks before opening any attachment. Any views or opinions presented are solely those of the author and do not necessarily represent those of Quadriga.

6 11

forums + portal for {lang}.centos.org sites
by Karanbir Singh 16 Aug '24

16 Aug '24

Hi, Earlier in the evening today Ralph, Fabian and I had a chat about the present state of the language subsites. This email sort of summarises the main issue ( s/w ). We seem to have run into a slight technical hitch with punbb/fluxbb. They dont support LDAP as a backend. And we had decided a few months back that all new rollouts must have ldap backend so we can rollin CentOS-DS / openldap based backend. So we need to look at alternatives, and since the primary focus of these international sites is going be forums : Here is a shortlist ( if there is anything else that people are aware of, please add to this list ) - phpBB - SMF - Fudforum - phorum - fluxbb Requirements: - Must be able to scale ( couple of hundred thousand msgs ) - Must be able to handle ldap auth ( if it cant, whats involved in writing the ldap-auth portion ) - Must address the specific requirements raised by the present www.centos.org forum users ( Can you please fill this section in ? ) - Must support all languages we need ( pure utf8 support would be good ) - Secure - Skin'able Nice to have: - Capable of running multiple instances from a single deployment - responsive community :D Things we will need to do: - Decide on what s/w to use. - Give the ArtWork people enough time to get the look & feel sorted. - Migrate newbb forums from www.centos.org to $system ( hey, english is a language too :D ). - Migrate fr.centos.org into the final s/w - setup {de/es/ja/it/pt_br}.centos.org Actions: Ralph and Fabian are going to work on setting up a test ldap server, once that is online we will then start by installing into our test-vm-farm the various s/w to eval them. If anyone would like to help, please feel free to jump right in. I'll setup a wiki page for this issue, which might be a good place to track progress. -- Karanbir Singh : http://www.karan.org/ : 2522219@icq

22 81

enhancing /etc/*-release
by Karanbir Singh 06 Apr '15

06 Apr '15

hi, We have spent a lot of time working out how best to work the /etc/centos-release file in order to satisfy most use cases. We've had a lot of positive feedback on the three digit release numbering that went into the first CentOS-7 release. And in the coming months, as the rolling builds onramp we will start seeing movement around those. This will also help address some points learned from the community during the 7.0.1406 cycle. We have also decided to split the /etc/redhat-release link to /etc/centos-release and use that as a way to better indicate what codebase the running CentOS Linux instance was derived from. Examples of what these files will look like in say March 2015 ( if .1 is released upstream by then ): ------------------- /etc/centos-release: CentOS Linux release 7.1.1503 (Core) /etc/redhat-release Derived from Red Hat Enterprise Linux 7.1 (Source) ------------------- The /etc/os-release file remains unchaged to indicate CentOS-7 as being the distro being consumed. We will however, be adding ABRT specific content to the os-release file once we have bugs.centos.org setup to accept abrt requests - and we have the required patched rolled out into the distro. These additions will have no impact on numbering reported via tools that consume /etc/os-release. The /etc/centos-release file will then evolve with every monthly release, with the updated file being pushed into updates repo. This implies if someone was to install from the March rolling build, their /etc/centos-release will already have 7.1.1503 and anyone having a previously installed machine, doing a yum update would see the same file drop in. The net result is an impact for new people installing from rolling build media ( and instance media like live images, cloud images, containers etc ). One installed or running, there is no change to how CentOS Linux has been in the past. You just get regular updates, and any machine, regardless of how it was installed and when, updated to the same point in time will have identical content. -- Karanbir Singh, Project Lead, The CentOS Project +44-207-0999389 | http://www.centos.org/ | twitter.com/CentOS GnuPG Key : http://www.karan.org/publickey.asc

14 27

CBS build tags and git branches names for SCL
by Honza Horak 17 Mar '15

17 Mar '15

I'm about to submit a couple of requests into bugs.c.o to create necessary infrastructure for collections from RHSCL-1.2 at least. But first I want to make sure about the tag and branches names. Currently we have some testing build tags: scl6-el6-mariadb100-build scl7-el7-mariadb100-build and branches in dist-git (in sclo project): scl-mariadb100-el6 scl-mariadb100-el7 And the SIG has codename 'sclo'.. So, it's kind of messy already in the beginning :) Thinking about keeping it simple and uniform, what about this way: Keep 'sclo' as SIG codename to indicate it's combination of SCL on openness :) CBS build tags in format like: sclo6-mariadb100-build sclo7-mariadb100-build and branches in dist-git (in both sclo and rpms projects): sclo6-mariadb100 sclo7-mariadb100 This mail is to check whether we're fine to go this way or if there are some limitations we need to follow. dist-git and CBS gurus, please, provide your PoV. Thanks! Honza

6 16

docker 1.5 in virt7-testing
by Karanbir Singh 13 Mar '15

13 Mar '15

hi guys, docker 1.5 is now in virt7-testing repos, please test and feedback so we can move to release.. thanks lokesh! - KB -- Karanbir Singh, Project Lead, The CentOS Project +44-207-0999389 | http://www.centos.org/ | twitter.com/CentOS GnuPG Key : http://www.karan.org/publickey.asc

3 3

cbs feature req: building embargoed content
by Karanbir Singh 12 Mar '15

12 Mar '15

hi guys, we need a way to build embargo'd content like reimzul's priv-build, that does not make stuff public till a specific date ( the actual push to public can be manual ); how would we achieve this on cbs ? I guess the git stuff is easier since it can just be a scratch build from srpm and people need not push the git content back to git.centos.org till embargo expiry. - KB -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

7 14

SIG repo layout and conflicts between projects within a SIG
by George Dunlap 10 Mar '15

10 Mar '15

At the moment, we're essentially using yum repos as the "patch" mechanism to CentOS Base: SIGs populate a repo with packages they want to add / override, and all other packages default to Base. The current setup is that every sig has exactly one production repo per CentOS version. Also, we seem to be wanting to see SIGs not as individual projects, but to make the 'G' an actual "group": so, for instance, the Virt SIG has both Xen and Docker, and we've recently approved oVirt. But there's a bit of a conflict here: what if one project in the SIG wants to use the package from Base, and another project in the SIG wants to use a custom-built package? Or what if both projects want to customize their packages in different ways? In particular: the Docker project in the Virt SIG wants to supply updated versions of Docker; but at the moment uses the kernel package from Base. The Xen project, on the other hand, needs to supply a different kernel package that has dom0 support enabled. Up until this point, this hasn't been an issue because I've been doing Xen development in virt6-testing, and Lokesh has been doing Docker development in virt7-testing. But I'm hoping to get Xen for CentOS 7 out soon, and I was just about to upload the Xen kernel to the CBS, when it occurred to me that if I did so, then everyone using the virt7-testing repo for Docker would suddenly get my updated kernel as well. So it seems we have a couple of ways to approach it. 1. Keep it one repo per SIG, and make all projects in a SIG sort out shared dependencies. 2. Allow one repo per *project* within a SIG (e.g., virt6-xen, virt7-docker) 3. Have some alternate way of doing a "patch", instead of repos. In the case of xen / docker, I haven't talked with Lokesh about this at all; it may be that he's fine with sharing a kernel with Xen. But it does increase the workload: now Lokesh needs to maintain the Docker side of the kernel, whereas before they could just rely on our upstream to do a lot of that testing. Furthermore, it increases the complexity of development and testing: before when doing kernel work, I could just test Xen; now the kernel needs to be tested by both Xen and Docker. And finally, now users of both projects get a kernel update whenever either one does. Which, of course, is pretty much the same as any other distro; but as a whole, at the moment at least, we have a lot less resources than any other distro our size and popularity. On the whole I'm inclined to prefer #2. Thoughts? -George

5 8

SCL and git 2.X
by Vacelet, Manuel 05 Mar '15

05 Mar '15

Hi all, I'm not sure I'm posting to the right place. If it's not I would be pleased if you could point me to the relevant place! TL;DR: I need recent version of git (2.3 ATM) and I need it on centos6 and centos7. For the time being, I managed to rebuild RH git19 scl for centos6 (last time I checked it was not already part of centos SCL but I might be wrong). My question is: is there other persons here interested by a recent, community supported, version of git ? and if yes, how can this be done (a SIG?) ? Cheers, Manuel -- Manuel VACELET Enalean Co-founder & Chief Technical Officer http://www.enalean.com Twitter @vaceletm <https://twitter.com/vaceletm>

6 8

Fwd: [Gluster-devel] OOM issue in openstack Cinder - GlusterFS CI env
by Justin Clift 27 Feb '15

27 Feb '15

Wondering if anyone in the CentOS Community has the time/interest to investigate this CentOS + Rackspace issue? Regards and best wishes, Justin Clift Begin forwarded message: > From: Deepak Shetty <dpkshetty(a)gmail.com> > Subject: Re: [Gluster-devel] OOM issue in openstack Cinder - GlusterFS CI env > Date: 26 February 2015 05:24:48 GMT > To: Niels de Vos <ndevos(a)redhat.com> > Cc: Gluster Devel <gluster-devel(a)gluster.org> > > A quick update > > By doing more experiments, we were able to figure that the failures are happening only on the rax (rackspace) cloud VMs, but not on the hpcloud VMs > Also the OOM issue happened on rax VMs even when gluster was not used as the storage (which means the cinder default storage uses LVMs) > > So right now its seems like a combination of rax & centos7 is causing the fail, making centos7 CI job unsuitable for openstack tempest runs. > Since openstack-infra jobs can be schedule across any cloud provider, we cannot use centos7 as the distro, until the rax+centos7 oom issue is resolved > > Given the strict cinder community guidelines for having a CI job in place, and openstack infra is having other issues with F20 and F21 based VMs, > (which are under debug as I was told on -infra IRC by infra core members ) the only option left is to use ubuntu as the base OS for running > cinder-gluster CI job, which we are in the process of evaluating and working on. > > > For full details refer to thread @ > http://thread.gmane.org/gmane.comp.cloud.openstack.devel/46861 > > For a quick summary, see this mail of the above thread @ > http://article.gmane.org/gmane.comp.cloud.openstack.devel/47287 > > > thanx, > deepak > > > On Mon, Feb 23, 2015 at 6:01 PM, Deepak Shetty <dpkshetty(a)gmail.com> wrote: > > > On Sun, Feb 22, 2015 at 8:46 PM, Niels de Vos <ndevos(a)redhat.com> wrote: > Hi Deepak and others, > > On Sat, Feb 21, 2015 at 10:00:18PM +0530, Deepak Shetty wrote: > > Hi All, > > I am looking for some help from glusterfs side for the Out of Memory > > (OOM) issue > > we are seeing when using GlusterFS as a storage backend for openstack > > Cinder (block storage service) > > This seems to be using the FUSE mount points for the VM images, right? > > Yes, no libgfapi. > > > > openstack has a upstream CI env managed by openstack infra team, where > > we added a new job that creates a devstack env (openstack all in one for > > newbies) and configures the block service (Cinder) with GlusterFS as > > storage backend. Once setup, the CI job runs openstack tempest (Integration > > test suite of openstack) that does API level testing of the whole openstack > > env. > > > > As part of that testing, ~1.5 to 2 hours into the run, the tempest job > > (VM) hits OOM and the kernel oom-killer kills the process with the max > > memory to reduce memory pressure. > > > > The tempest job is based on CentOS 7 and uses glusterfs 3.6.2 as the > > storage backend for openstack Cinder > > > > The openstack-dev thread @ > > *http://thread.gmane.org/gmane.comp.cloud.openstack.devel/46861 > > <http://thread.gmane.org/gmane.comp.cloud.openstack.devel/46861>* has > > details including links to the logs captured from the tempest jobs. > > http://fungi.yuggoth.org/tmp/logs2.tar seems to be the most useful. > > > Per the openstack infra folks, they have other jobs based off CentOS7 that > > doesn't hit this issue, the only change we are adding is configuring cinder > > with glusterfs when this happens, so right now glusterfs is in the > > spotlight for causing this. > > Well, there definitely is something wrong on the Gluster side. After the > 'blackout' between 17:47 - 18:17, this is logged in > logs/screen-c-vol.txt.gz: > > mount -t glusterfs 127.0.0.1:/vol1 /opt/stack/data/cinder/mnt/f4c6ad7e3bba4ad1195b3b538efab64a" returned: 32 > > 32 as exit code from "mount" means "mount failure". > > Yeah, but thats the time when things went bad in the system leading to OOM > So can't we safely assume that system was in a bad state, hence mount failure... > > We actually need to look before 17:47 and at/around 17:47 to figure what was running that caused the > blackout to happen. > > I do see some errors I see in the brick logs and warnings in the mount logs > but I Don't think they are anyway related to or cause of creatign OOM. > > > > The mountpoint would log details in > logs/glusterfs/opt-stack-data-cinder-mnt-f4c6ad7e3bba4ad1195b3b538efab64a.log. > At 18:00 in that file, there is this message: > > 0-vol1-client-0: server 104.130.30.179:49152 has not responded in the last 42 seconds, disconnecting. > > Its a one-brick volume, so this error is pretty fatal. This error does > not cause an unmount, but the mountpoint will return errors. Subsequent > mount tries will fail, because the mountpoint is not unmounted yet. > Tests to see if the target directory can be used to mount a volume, will > fail, causing the mount process to return error 32. > > Thats in the middle of the blackout, so disconnections are ok, i feel. > > > > A little later, the (still) existing fuse mount client manages to > connect to the brick again: > > 0-vol1-client-0: changing port to 49152 (from 0) > > The log of the brick does not contain anything useful for this time :-/ > > > I am looking thru the logs trying to co-relate syslog, dstat, tempest info > > to figure the state of the system and what was happening at and before the > > OOM to get any clues, but wanted to start this thread in gluster-devel to > > see if others can pitch in with their ideas to accelerate the debug and > > help root cause. > > I have not yet found any pointer to what could have taken place between > 17:47-18:17. It seems that network I/O stopped (ping timeout in FUSE). > > During that tiem from the tempest logs i could figure that among other tests > VolumesBackupV1Test was running, which reads from Cinder- Gluster backend > and write to swift service, which will add good IO traffic. > > > One of the few logs that I came across which contains some messages > during this 'blackout', is logs/screen-n-cond.txt.gz. But I assume these > messages are triggered by the underlying network issues. > > > You mention that there was a OOM? Which logs contain some details about > that? Also, could you explain what test (workload) was running in the > VM(s?) on the FUSE mountpoint? > > Look at logs/syslog.txt (thats /var/log/messages), grep for oom-killer > so at 18:17 when it happend it killed nova-compute and mysqld as they were the > most memory hoggers. But in talking with openstack infra folks, they indicated > that they have similar CI jobs where mysqld doesn't get killed, so adding gluster to > the mix is definitely the reason for reaching OOM state. > > There is really no application running workload here. Most of the IO is generated as part > of oepnstack orchestration tests, liek backup (which reads from cinder and writes to swift), clone > (create a snap and then copy it), create/delete cinder volumes (which are files in glusterfs) etc > > From what i understood in talking with the openstack infra folks they referred me to > the dstat logs sayign the kernel buffer cache is going as low as double digit M and system is in > 70-90 % cpu wait time, which to them indicates that the IO of the system has gone bonkers > > Below is the exact thing from the chat log (for the 1st re-create / logs) : > > Feb 20 21:46:28 <sdague> deepakcs: you are at 70% wait time at the end of that > > Feb 20 21:46:37 <sdague> so your io system is just gone bonkers > > Feb 20 21:47:14 <fungi> sdague: that would explain why the console login prompt and ssh daemon both stopped working, and the df loop in had going in my second ssh session hung around the same time > Feb 20 21:47:26 <sdague> yeh, dstat even says it's skipping ticks there > Feb 20 21:47:29 <sdague> for that reason > > Feb 20 21:47:48 <fungi> likely complete i/o starvation for an extended period at around that timeframe > Feb 20 21:48:05 <fungi> that would also definitely cause jenkins to give up on the worker if it persisted for very long at all > > Feb 20 21:48:09 <sdague> yeh, cached memory is down to double digit M > > > Feb 20 21:49:19 <fungi> because that's 00% of what it entails, at least for the hard problems > Feb 20 21:49:21 <sdague> deepakcs: so, honestly, what it means to me is that glusterfs is may be too inefficient to function in this environment > Feb 20 21:49:25 <fungi> er, 99% > Feb 20 21:49:34 <sdague> because it's kind of a constrained environment > > Feb 20 21:51:18 <deepakcs> sdague, skipping ticks means ? > > Feb 20 21:51:32 <sdague> deepakcs: dstat runs once a second > Feb 20 21:51:38 <sdague> unless the system gets under preasure > Feb 20 21:51:54 <sdague> then it starts skipping it's runs to prevent things from getting worse > Feb 20 21:52:01 <sdague> i.e. to not be part of the problem > > Feb 20 21:53:24 <dansmith> deepakcs: dstat is a well-understood thing, so you can probably figure out all you need to know from the docs > > Feb 20 21:53:41 <dansmith> deepakcs: it's run in the jobs so we can identify issues like this where the system is overly taxed and unable to actually run the tests > Feb 20 21:53:57 <deepakcs> dansmith, agreed, just trying to understand system debug better :) > > Feb 20 21:54:39 <dansmith> deepakcs: http://linux.die.net/man/1/dstat > Feb 20 21:54:54 <fungi> deepakcs: i have another manual run started just now, with the devstack-gate timeout cranked up to 5 hours just to make sure we capture the full job duration > > Feb 20 21:55:37 <deepakcs> dansmith, thanks :) > Feb 20 21:55:41 <fungi> deepakcs: i'll keep you posted in here or follow up to the e-mail thread if you're gone > > Feb 20 21:57:42 <deepakcs> sdague, dansmith what i am not clear is that mysqld was the one with the max mem taken, which is why oom killer killed it, so howz gluster connected to creating oom here ? > Feb 20 21:58:19 <dansmith> deepakcs: buffer cache can create huge amounts of memory pressure > Feb 20 21:58:34 <dansmith> deepakcs: the largest memory user gets the boot just because it's the most likely to reduce the pressure > > Feb 20 21:58:48 <dansmith> deepakcs: there is a good data point here that mysql doesn't get killed in the other jobs > > Feb 20 21:59:20 <deepakcs> dansmith, ok, i looked up the glusterfsd (the process doing the IO to the disk) and its max 1.7G ram > Feb 20 21:59:21 <dansmith> deepakcs: nor do I know what's going on in your job > Feb 20 21:59:31 <dansmith> deepakcs: that's user memory > Feb 20 21:59:49 <deepakcs> dansmith, yes and glusterfs is a user space FS entirely > Feb 20 22:00:03 <dansmith> deepakcs: is it writing with O_DIRECT? > > Feb 20 22:00:38 <deepakcs> dansmith, the logs doesn't have glusterfs logs, so cant tell, but we don't configure it w/ o_direct by default, so it shouldn't > Feb 20 22:00:55 <dansmith> deepakcs: so if not, every write it makes generates buffer cache load in the kernel > > Feb 20 22:01:24 <dansmith> deepakcs: regardless, this is really not the right forum to discuss this level of system debug I think > Feb 20 22:01:35 <deepakcs> dansmith, ok > Feb 20 22:01:59 <dansmith> deepakcs: the stats runs from your job and other non-failing jobs should be helpful, and it's clear that something in your job is eating all the resources > > > > > In logs/screen-dstat.txt.gz there seems to be a process called > "cinder-backup" that is running around that time. From the details of > dstat, I guess that this process causes the reading of files from the > Gluster Volume. When the system becomes responsive again, dstat has 11G > as "dsk/total read", it jumped from ~400M in 30 minutes. If this amount > of data is not flushed but kept in memory, I don't doubt that an OOM > could be triggered. > > You do or don't doubt ? - thats confusing :) > > Yes, I am also doubting that IO traffic generated by doing backup tests is the only scenario > where most IO is generated (from Cinder-GlusterFS perspective). I looked at all the tests > that openstack Cinder has, and this was the only intensive test. > > Given that Backup was running before the time system went OOM, it seems liek a good > one to doubt upon. > > Plan is to disable Backup tests and see if we still hit the OOM issue. > > > > Do you know where this backup is written to? If we can setup a > reproducer environment that emulates the workload, things would become > much easier for debugging :) > > In the current setup, backup source is GlusterFS and target is swift (configured to > store its data in /opt/stack/swift/ mounted on 2nd disk) > > > One of the 1st things I would try, is to configure the kernel to flush > dirty memory more aggressively ('sysctl' values for 'vm.*' options). An > other thing that may help to identify the memory usage, is capturing the > output of 'slabtop --once' in a 1-minute interval. > > Unfortunately, openstack infra doesn't provide that kind of job control. The idea is to have a uniform > env for all storage backends and just change the specific backend that one wants to test, so that the runs are > comparable agaisnt a common base. Atleast that my understanding. > > But yes there are knobs given to tweak/tune the kind of tests you wanna run, which is what I plan > to exploit and turn off Backup tests and see if it helps keep system in sane state. > > Thanks Neils for your time and analysis, it really helped. > > I m hoping now with the backup tests disabled, we won't hit, will update here once openstack infra folks co-operate > to run the CI job manually for us! > > > thanx, > deepak > > > _______________________________________________ > Gluster-devel mailing list > Gluster-devel(a)gluster.org > http://www.gluster.org/mailman/listinfo/gluster-devel -- GlusterFS - http://www.gluster.org An open source, distributed file system scaling to several petabytes, and handling thousands of clients. My personal twitter: twitter.com/realjustinclift

2 2

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel February 2015