virt November 2014

virt@lists.centos.org

7 participants
6 discussions

Virt SIG meeting minutes
by George Dunlap 16 Dec '14

16 Dec '14

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Attendees: Lars, Lokesh, George, KB (eventually) Docker update: - - Docker 1.3.1 for virt7-testing Xen update: - Xen 4.4.1 in virt6-testing - Can't do a release of xen without libvirt because libvirt depends on a specific version of libxl - Plan: update to libvirt-1.2.9 before test then release Signing - C7 we want to use the CBS signing system which isn't up yet - All SIGs will just wait to release C7 stuff until then - x4c C6 is signed with distro key, so we'll keep having Johnny sign those Target: C6 xen/libvirt beta release in mid-November, release early December C7 docker: release when sig ready C7 xen: beta maybe mid-December Mid-December: Dojo schedule - Biggest one: US in early Summer - 30 Jan before FOSDEM - ~300 people - Sponsorship? - Talk / workshop / hack session re virt sig Pre-FOSDEM dojo: - kbsingh: It would be good to have virt-sig presence at dojo before FOSDEM - gwd: I may not be able to make it; I'll try to find someone to cover me. IRC chat log: <gwd> Virt SIG meeting in 2 minutes *** jonludlam (~jonludlam(a)guest-gw.chu.cam.ac.uk) has joined channel #centos-devel [13:59] *** tazz (~gaurav(a)triband-mum-120.60.128.133.mtnl.net.in) has joined channel #centos-devel *** dominic_ (~dgeevarg(a)59.93.70.149) has quit: Quit: Leaving <pasik> hello [14:00] <gwd> OK, who have we got present? I see pasik and jonludlam [14:02] <gwd> kbsingh: You around? <gwd> lsm5? [14:04] <gwd> OK, well I guess let's get started. [14:05] *** agrimm (agrimm@nat/redhat/x-lypwfoeufylzzmdt) has joined channel #centos-devel <jonludlam> gwd, is there an agenda? [14:06] <gwd> The main agenda was updates: update on Xen, on Docker, and on the CBS. <gwd> The CBS is the main blocker for a Docker release <gwd> And really for me doing anything useful as far as a Xen update as well. [14:07] <jonludlam> I still don't have my cbs account <gwd> And quaid had asked about oVirt joining the Virt SIG as well <gwd> jonludlam: Right, I didn't realize you'd asked for one. [14:08] * gwd tries to remember the nick of the guy who was working on that... <gwd> Well the Xen status is: <gwd> * I've got Xen 4.4.1 built in the CBS (unsigned) <gwd> * It's using an "updated" blktap2.5, but I'm not sure how stable it is (needs more testind) [14:09] <jonludlam> (tbh I hadn't been pinging about it as I've had to work on other things recently...) <gwd> * And before we can do a proper release, I'd need to rebuild libvirt <jonludlam> gwd, is it using the dkms blktap module? <jzb> Evolution: ping <Evolution> jzb: pong [14:10] <gwd> I think the kernel probably is, not positive. <jzb> Evolution: where can I find minutes from last week's meeting? <jzb> Evolution: Atomic SIG meeting, I mean, in case that was vague. :-) <gwd> jonludlam: Any reason for asking? :-) <Evolution> everything should be at -> http://www.centos.org/minutes/ [14:11] <jonludlam> gwd, the dkms module is quite old, and I've had issues using it under ubuntu. The newer kernel module source can't be built out of tree though <gwd> Yes, the whole blktap thing is something which needs to be sorted out. [14:12] <jonludlam> given we're providing a new kernel, it might make sense to have the in-tree patches in the kernel <jonludlam> gwd, totally agree <pasik> i think the current linux 3.10 kernel in xen4c6 is based on some citrix patches <pasik> for blktap2.5 [14:13] <gwd> That would make sense -- I haven't had time to actually get to look at the kernel package and see what's in it. <pasik> i remember some discussions/emails with David Vrabel about those patches <jzb> Evolution: found, thx <jonludlam> pasik, might be worth refreshing the patches - we've done a lot of testing on them recently [14:14] <pasik> jonludlam: ok, good to know <pasik> jonludlam: can you point to a git tree with the latest patches? <jonludlam> pasik, not sure - I'll have a look <pasik> jonludlam: thanks! [14:15] <jonludlam> https://github.com/xenserver/linux-3.x.pg/tree/master/master I believe *** telex (teletype(a)freeshell.de) has quit: Remote host closed the connection [14:17] <gwd> pasik: You haven't tested the 4.4.1 patches in virt6-testing, have you? [14:18] *** telex (teletype(a)freeshell.de) has joined channel #centos-devel <kbsingh> gwd: here now <kbsingh> +1 for prepatching the kernel for needed support, dkms is never worked for me in prod [14:20] <gwd> kbsingh: OK -- I've just done a status update for Xen (basically not much update since last time, except that I found a pretty nasty bug upstream and fixed it) <kbsingh> on the CBS update : [14:21] <pasik> gwd: hmm, i *think* i did test the rpms you linked <pasik> gwd: but i'm not totally sure. i've been way too busy with other things <pasik> gwd: i'll test again to be sure :) <gwd> pasik: Sure. :_) <kbsingh> lots of progress, MerlinTHP is working on closing up the lookaside binary sources import script, and I am working on the gitblit update for git.centos.org - having those two things done will allow prettymuch the entire workflow to move from srpms to git <lsm5> gwd: yup [14:22] <pasik> kbsingh: any eta? <kbsingh> w.r.t signing packages, we have a sign process setup, need to sync with the koji guys and workout a release process. <kbsingh> pasik: 2 weeks is a good eta to plan against <gwd> C6 isn't going to use the CBS signing anyway, since the existing C6 packages use the CentOS signing key. <lsm5> sorry about the delay, Re: docker, we have 1.3.1 up, nothing else to update :\ [14:23] <kbsingh> we'll likely have chunk of the work done for the next buildsys meeting on Monday coming, and i'd give it another week for people to break i <kbsingh> gwd: if we can find a migration path, would you prefer it went through cbs ( c6 ) <lsm5> lsm5: and i'm back at RH if that matters :) *** chorrell (~chorrell(a)72.15.63.49) has joined channel #centos-devel <kbsingh> lsm5: we;ll try not holding that against you <lsm5> :) [14:24] <gwd> kbsingh: That would be fine with me, but I don't know any of the details there... *** arielb (~arielb@fedora/arielb) has joined channel #centos-devel <kbsingh> gwd: ok, maybe we can thrash out the detalis on the virt-list. it will involve shipping a xen that requires: centos-release-xen <version>; which then has the new key <kbsingh> and have the centos-release-xen conflict with the older xen <kbsingh> gwd: in terms of targets, are we looking for a 4.4 into c6 or c7 now ? [14:26] <kbsingh> as in, what are we doing first ? *** BRLX (~Thunderbi(a)rtr.ak-p.at) has quit: Quit: BRLX [14:27] <gwd> kbsingh: Well I thought changing one thing at a time would be good. :-) So I'm aiming at 4.4 in c6 <kbsingh> cool <gwd> Also, updating to 4.4 in c6 doesn't require kernel changes; just xen & libvirt packages *** BRLX (~Thunderbi(a)rtr.ak-p.at) has joined channel #centos-devel <kbsingh> second question i had was for jonludlam i guess, there was talk of updating to a 3.14 kernel, is that still on the cards ? to better align with upstream expectations [14:28] <gwd> kbsingh: Do we want to update to 3.14 in c6? We were definitely planning on doing a kernel update for c7 <kbsingh> if there is a good reason to, we can update in c6 as well as c7 - specially if its going to mean only having to maintain one kernel tree [14:29] <jonludlam> I have no particular opinion on kernel version <jonludlam> David V might though [14:30] <gwd> jonludlam: I can ask the Linux Xen crowd what they think is a good option <kbsingh> maybe something to thrash out a bit then, for now for 4.4 into c6, if its not a blocker then lets not bring it to the agenda [14:31] <jonludlam> gwd, sounds like a good idea <kbsingh> col [14:32] <kbsingh> cool even <gwd> OK, so the plan is: First c6 xen+libvirt; then possibly c6 kernel update; then c7 everything. <gwd> Anything else for Xen? <pasik> libvirt status? [14:33] <pasik> for xen4.4 <gwd> pasik: I was planning on updating it to whatever the most recent release. <gwd> But I haven't had a chance to look at it. <pasik> ok [14:35] <pasik> libvirt obviously needs to be built and tested before we can push the xen 4.4 update to c6 [14:36] <gwd> Yes, that's basically the delay at this point... *** smerrill-offline (~smerrill(a)ec2-174-129-193-80.compute-1.amazonaws.com) is now known as smerrill <gwd> lsm5: Anything you're blocked on for a c7 docker release, other than the cbs being set up? [14:37] <kbsingh> do we want to include the upstream libvirt people as well ? <kbsingh> DV: ping <lsm5> gwd: no blockers <gwd> lsm5: Cool. *** mbrysa (mbrysa@nat/redhat/x-toigkxrcipgtvkod) has quit: Quit: Leaving [14:39] <gwd> quaid also mentioned oVirt potentially wanting to do join the Virt SIG; I presume to have a re-build of qemu with some extra features enabled <gwd> Is there anything to discuss about that? *** blip (~dewey@unaffiliated/blip) has quit: Ping timeout: 258 seconds <gwd> kbsingh: Also, I have in the last meeting minutes something about dojos and sponsorship? Was there an action item there for anyone? [14:40] <kbsingh> unsure, but there are a few ovirt people on the virt-list now, so we can ping there *** mattgriffin (~textual(a)99-181-54-51.lightspeed.rcsntx.sbcglobal.net) has joined channel #centos-devel <kbsingh> gwd: we have a dojo on the day before Fosdem, I think virt-sig should do something there, either a hack session or a talk session <gwd> kbsingh: quaid responded to my e-mail announcing the Virt SIG meeting on centos-virt <gwd> Hrm <gwd> fosdem> It's looking somewhat likely that I won't be able to make it to FOSDEM next year, due to some personal stuff I have to sort out in the US around that time. [14:41] <kbsingh> gwd: ok, we should then try and find someone who can cover ? [14:43] <kbsingh> brb <kbsingh> back [14:44] <gwd> kbsingh: Yes, that would be good. [14:45] <gwd> kbsingh: Let me see if I can find someone to get involved. [14:46] <kbsingh> thanks *** chandankumar (chandankum@nat/redhat/x-kxxugjygxkcyqqlh) has quit: Quit: Leaving <gwd> OK, anything else to cover? <kbsingh> the cloudsig is starting to finally get off the ground, do we want to cross polinate with them a bit ? [14:47] <kbsingh> maybe have someone from virtsig hangout in their meetings and ask them to come along here as well <pasik> jonludlam: would you like to email about the blktap2.5 kernel patches once you figure it out? <gwd> kbsingh: They're doing stuff like openstack, right? <jonludlam> pasik, if there's anything actually to do, I will [14:48] <kbsingh> gwd: yeah, openstack and eucalyptus looks to be what they are starting with, then cloudstack and opennebula miht comealong later. <jonludlam> it looks like the recent activity on the patch in our tree has been to do with very large scalability <jonludlam> which might not be of much interest <pasik> jonludlam: thanks! [14:49] <pasik> jonludlam: well, at least all the bugfixes are of interest :) <jonludlam> pasik, for sure :-) [14:50] <gwd> kbsingh: Yes. I'll see if I can start lurking on their meetings. <jonludlam> talking of other SIGs, what needs to happen to kick off the ocaml SIG? [14:51] <gwd> Just real quick -- any other business we need to cover before we close the Virt SIG meeting? <kbsingh> jonludlam: not much, we just need to find a couple of problem definitions that sig will try and solve, then propose it and bob's your uncle *** lvv (~loganaden(a)RNT-ENG2.dhcp.mu.afrinic.net) has quit: Quit: lvv <kbsingh> gwd: not from me. <jonludlam> is 'we need it for xapi' a sufficient problem definiton? [14:52] <jonludlam> I'm sure there are more problems it solve too... :-) [14:53] <kbsingh> jonludlam: thats a good place to start from [14:54] <kbsingh> jonludlam: lets pickup on that over email, i need to rebase over to another call <jonludlam> sure *** santana (9f99820a(a)gateway/web/freenode/ip.159.153.130.10) has joined channel #centos-devel <gwd> Great, thanks everyone! [14:55] -----BEGIN PGP SIGNATURE----- Comment: GPGTools - https://gpgtools.org iQEcBAEBCgAGBQJUa15LAAoJELIVx6fHhBvt6R0H/29i5r7sTPChWEa6r4d7LfC9 /nkEwTP/K/eYmcfLy5tZZr09dQWyrSoQnbVK4mudTiy7s0+I4EEj8+8bSTCDWuNV CgUUB0vP18O2Zw7VMbevaM64MgmFNn/TwHyOOIJyiXQZWAdPVMfDHpz2chnV3jzC aS1E16RPHEuxppnteZXmfdbRrXQ+CM3uI5AodpHDvk1kN10OciHjNXknk8ok82nz AGrgH0BmxYyu+BxRgAqivIO65YlIQvLIKHwphElY33rXTFkucH59o6gMmTWOR4DE Y5olalxIrFvIHEWDh8BY/Jbytvewf3KO0Jpw+2YBHi2xzDAzYhJyXPPb/z3/hnY= =5z5s -----END PGP SIGNATURE-----

2 1

Problem with Xen4CentOS
by Thomas Weyergraf 18 Nov '14

18 Nov '14

Hi folks, we (the company i am working for) are running several dozens of virtualisation servers using CentOS 6 + Xen4CentOS as the virtualisation infrastructure. With the latest versions of all packages installed [1], we see failures in live-migration of stock-CentOS6 HVM guests, leaving a "Domain-unnamed" on the source host, while the migrated guest runs fine on the target host. Domain-0 0 2048 64 r----- 6791.5 Domain-Unnamed 1 4099 4 --ps-- 94.8 The failure is not consistently reproducable, some guests (of the same type) live-migrate just fine, until eventually some seemingly random guest fails, leaving a "Domain-unnamed" Zombie. That Domain-unnamed causes several problems: - The memory allocated to Domain-unnamed remains blocked, thus creating a veritable 'memory-leak' to the host - The DomU causing Domain-unnamed cannot be restarted on the host, as xm thinks it's already running I have tried various things to get rid of Domain-unnamed, all without success - multiple xm destroy - restart xend - delete everything regarding Dommain-unnamed in xenstore with xenstore-rm. The removal is successful, but the domain remains. Restarting xend after the deletion restores Domain-unnamed in xenstore So far, the only way to get rid of Domain-unnamed is a virt-host reboot. As these hosts are all quad-socket opteron 6272 machines with 256gig ram running dozens of guests, this is highly impractical. I have seen this behaviour using xen 4.2.5. The previous 4.2.4 versions did not show this problem, however we did not use live-migration extensively prior to that. Before switching to Xen4CentOS, we used to build our own Xen 4.2.2 based on a git repo, published by Karanbir Singh. We had several issues with that version, but never observed a "Domain-unnamed". Any idea how to resolve this issue would be highly appreciated, as working live-migration is crucial to us. Regards, Thomas Weyergraf Some notes on our config: 1. We still use xm/xend for various reasons ---- 2. Our grub-config for the virtx-hosts is as follows: ---- default=0 timeout=5 #splashimage=(hd0,0)/grub/splash.xpm.gz #hiddenmenu title CentOS (xen-4.2.5-37.el6.gz vmlinuz-3.10.56-11.el6.centos.alt.x86_64) root (hd0,0) kernel /xen-4.2.5-37.el6.gz iommu=1 console=vga,com1 com1=115200,8n1 vga=text-80x25 dom0_mem=2048M,max:2048M module /vmlinuz-3.10.56-11.el6.centos.alt.x86_64 ro xencons=hvc0 console=hvc0 root=/dev/fravirtx68/root rd_NO_LUKS LANG=en_US.UTF-8 KEYBOARDTYPE=pc KEYTABLE=de-latin1-nodeadkeys rd_NO_MD SYSFONT=latarcyrheb-sun16 rd_LVM_LV=fravirtx68/root rd_NO_DM module /initramfs-3.10.56-11.el6.centos.alt.x86_64.img title CentOS (vmlinuz-3.10.56-11.el6.centos.alt.x86_64) root (hd0,0) kernel /vmlinuz-3.10.56-11.el6.centos.alt.x86_64 ro root=/dev/fravirtx68/root rd_NO_LUKS LANG=en_US.UTF-8 KEYBOARDTYPE=pc KEYTABLE=de-latin1-nodeadkeys rd_NO_MD SYSFONT=latarcyrheb-sun16 rd_LVM_LV=fravirtx68/root rd_NO_DM module /initramfs-3.10.56-11.el6.centos.alt.x86_64.img 3. A typical guest-config looks like: ---- name = "fraappmgmt05t.test.fra.net-m.internal" uuid = "3778a443-9194-4c46-adff-211d7fcc24da" memory = "4096" vcpus = 4 kernel = "hvmloader" builder = 'hvm' disk = [ 'phy:/dev/disk/by-path/ip-192.168.240.7:3260-iscsi-iqn.1992-08.com.netapp:navfiler21-lun-56,xvda,w', 'phy:/dev/disk/by-path/ip-192.168.240.7:3260-iscsi-iqn.1992-08.com.netapp:navfiler21-lun-57,xvdb,w', ] vif = [ 'mac=00:16:3e:fa:15:4a,bridge=xenbr11' ] device_model = 'qemu-dm' serial='pty' xen_platform_pci=1 on_poweroff = "destroy" on_crash = "restart" 4. The xend.log excerpt of the migration process from the source host: ---- [2014-11-06 22:53:01 13499] DEBUG (XendDomainInfo:1795) Storing domain details: {'console/port': '7', 'cpu/3/availability': 'online', 'description': '', 'console/limit': '1048576', 'cpu/2/availability': 'online', 'vm': '/vm/f5139575-984b-4c28-b470-efc042ba2703', 'domid': '1', 'store/port': '6', 'console/type': 'ioemu', 'cpu/0/availability': 'online', 'memory/target': '4194304', 'control/platform-feature-multiprocessor-suspend': '1', 'store/ring-ref': '1044476', 'cpu/1/availability': 'online', 'control/platform-feature-xs_reset_watches': '1', 'image/suspend-cancel': '1', 'name': 'migrating-fraapppeccon06.fra.net-m.internal'} [2014-11-06 22:53:01 13499] INFO (XendCheckpoint:423) xc_save: failed to get the suspend evtchn port [2014-11-06 22:53:01 13499] INFO (XendCheckpoint:423) [2014-11-06 22:53:34 13499] DEBUG (XendCheckpoint:394) suspend [2014-11-06 22:53:34 13499] DEBUG (XendCheckpoint:127) In saveInputHandler suspend [2014-11-06 22:53:34 13499] DEBUG (XendCheckpoint:129) Suspending 1 ... [2014-11-06 22:53:34 13499] DEBUG (XendDomainInfo:524) XendDomainInfo.shutdown(suspend) [2014-11-06 22:53:34 13499] DEBUG (XendDomainInfo:1882) XendDomainInfo.handleShutdownWatch [2014-11-06 22:53:34 13499] DEBUG (XendDomainInfo:1882) XendDomainInfo.handleShutdownWatch [2014-11-06 22:53:34 13499] INFO (XendDomainInfo:2079) Domain has shutdown: name=migrating-fraapppeccon06.fra.net-m.internal id=1 reason=suspend. [2014-11-06 22:53:34 13499] INFO (XendCheckpoint:135) Domain 1 suspended. [2014-11-06 22:53:35 13499] INFO (image:542) signalDeviceModel:restore dm state to running [2014-11-06 22:53:35 13499] DEBUG (XendCheckpoint:144) Written done [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:3077) XendDomainInfo.destroy: domid=1 [2014-11-06 22:53:35 13499] ERROR (XendDomainInfo:3091) XendDomainInfo.destroy: domain destruction failed. Traceback (most recent call last): File "/usr/lib64/python2.6/site-packages/xen/xend/XendDomainInfo.py", line 3086, in destroy xc.domain_destroy(self.domid) Error: (16, 'Device or resource busy') [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2402) Destroying device model [2014-11-06 22:53:35 13499] INFO (image:619) migrating-fraapppeccon06.fra.net-m.internal device model terminated [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2409) Releasing devices [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2415) Removing vif/0 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:1276) XendDomainInfo.destroyDevice: deviceClass = vif, device = vif/0 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2415) Removing console/0 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:1276) XendDomainInfo.destroyDevice: deviceClass = console, device = console/0 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2415) Removing vbd/51712 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:1276) XendDomainInfo.destroyDevice: deviceClass = vbd, device = vbd/51712 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:2415) Removing vbd/51728 [2014-11-06 22:53:35 13499] DEBUG (XendDomainInfo:1276) XendDomainInfo.destroyDevice: deviceClass = vbd, device = vbd/51728 [2014-11-06 22:53:36 13499] DEBUG (XendCheckpoint:124) [xc_save]: /usr/lib/xen/bin/xc_save 26 2 0 0 5

2 3

Reminder: Virt SIG meeting on #centos-devel tomorrow, 2pm GMT
by George Dunlap 17 Nov '14

17 Nov '14

Just a reminder everyone, we'll be having our biweekly Virt SIG meeting Tuesday at 2pm GMT, on IRC, channel #centos-devel on freenode. If you have any particular items you'd like to put on the agenda, please let me know. Lars will be on a plane to China at that point, so I'll be chairing the meeting. -George

2 1

Re: [CentOS-virt] Official AWS AMIs: eu-central-1 support?
by Paul Schyska 14 Nov '14

14 Nov '14

Are the build scripts for the AMIs available somewhere? At least we could build them ourselves. Alternativly, maybe someone of the cloud group can build an AMI and share it as public AMI? We want to use Centos 7 too, and a migrated Image of mine refuses to start in eu-central-1. Thanks, Paul Schyska

1 0

Official AWS AMIs: eu-central-1 support?
by Axel S 13 Nov '14

13 Nov '14

Hello all, and thank you very much for the Centos 7 HVM AMI [1], greatly appreciated! This AMI does not work on the newly launched AWS Frankfurt region (eu-central-1). Any plans for fixing that? According to the AWS announcement [2], "The Region supports all sizes of T2, M3, C3, R3, and I2 instances". (We'll be running the CentOS 7 HVM AMI going forward, but looks like none of the other AMIs works on eu-central-1 either.) [1] https://aws.amazon.com/marketplace/pp/B00O7WM7QW [2] http://aws.amazon.com/blogs/aws/aws-region-germany/ Best, Axel

2 2

Xen Project Test Day for 4.5 RC2 on Thursday, Nov 13
by Russell Pavlicek 11 Nov '14

11 Nov '14

Since the folks on this list work with integrating Xen Project on CentOS, we'd like you to be aware of the upcoming Xen Project 4.5 RC2 Test Day. This Thursday, November 13, is our second Test Day for the 4.5 release cycle. Release Candidate 2 is now available for assessment. Information about testing this release can be found here: http://wiki.xenproject.org/wiki/Xen_4.5_RC2_test_instructions To learn more about Test Days, check out: http://wiki.xenproject.org/wiki/Xen_Project_Test_Days Russ Pavlicek Xen Project Evangelist, Citrix Systems Home Office: +1-301-829-5327 Mobile: +1-240-397-0199 UK VoIP: +44 1223 852 894

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

virt November 2014