- Devel - oVirt List Archives

Change functionality of some DB common functions
by Eli Mesika 21 Sep '14

21 Sep '14

Hi Up to now , using some common DB functions of delete/rename/change type of column were exiting silently if you had a typo in the table/column name We had a DWH patch made by Shirly that had changed column type for several columns but one of the changes had a typo in the column name This was only found bu QE and potentially could passed QE and found only in production. I had changes those functions not to allow such cases in patch http://gerrit.ovirt.org/#/c/32959 I think that this is a minor change for developers that will get exceptions now for such mistakes , but the benefit for production is clear and obvious to me. I will be glad to get your feedback on this patch Thanks Eli Mesika

2 1

Propose Shirly Radco as DWH & Reports Maintainer.
by Yaniv Dary 20 Sep '14

20 Sep '14

Hi, Shirly Radco has contributed many features, bug fixes and documentation enhancements for DWH & Reports and has taken a major role in the project progress and management in the last 9 months. I would like to propose Shirly as a DWH & Reports projects Maintainer. I would like to thank Shirly for her great contribution and hope she will keep up the good work! --- Yaniv Dary BI Software Engineer Red Hat Israel Ltd. 34 Jerusalem Road Building A, 4th floor Ra'anana, Israel 4350109 Tel : +972 (9) 7692306 8272306 Email: ydary(a)redhat.com IRC : Yaniv D

5 4

bug 1041569 question was: ovirt-d] [QE][ACTION REQUIRED] oVirt 3.5.0 status (fwd)
by R P Herrold 20 Sep '14

20 Sep '14

I was running down the open ovirt bugs, and notice this one, which has an outlink to a post by you (Eric) on a mailing lsit last month > And the following dependencies still open: > Bug 1041569 - [NFR] libvirt: Returning the watermark for all > the images opened for writing https://www.redhat.com/archives/libvir-list/2014-August/msg00207.html wherein you refer to a 'watermark'. To me as a non-lingo aware (as to libvirt / libguestfs) developer, a watermark is a pattern overlaid onto an image (a pattern placed on the glue side of a postage stamp, or a 'DRAFT" notice on a document) I _think_ you are referring to a 'high water mark' limit which, if exceeded would result in content being lost (there is mention of ENOSPC) Is this accurate? If so, isn't this information filesystem specific and so something which the outsider 'container' cannot and should be expected to reliably know? (I am thinking here of inode exhaustion even when a Linux type FS seemingly has space per: du ; the problem is worse with a 'sparse' capable filesystem in play; and seems wholly unanswerable with a Windows FS or some other variant) I don't see how the hypervisor could accurately ever know this without 'cracking open' and asking the inside tenant .. and for privacy and data integrity / security reasons, the hypervisor should not ever be doing this Thanks -- Russ herrold

2 1

OVIRT-3.5-TEST-DAY-3: NUMA support
by Greg Padgett 19 Sep '14

19 Sep '14

Hi, I was slated to test NUMA support along with Adam Litke. As we had to borrow a NUMA-enabled host (thanks Gilad) to perform the tests, I coordinated with Adam to get the testing done and ended up using the host most of the day. Tests performed: Basic sanity test: create/start/stop/destroy VMs with varying NUMA configurations. The running VMs appeared to have the correct NUMA node assignments (presuming the libvirt xml was respected), and everything worked as expected. REST API: GET operations succeeded and appeared correct for the various NUMA-related VM and Host resources. I didn't get the opportunity to perform the POST/PUT/DELETE operations. UI: I built the engine and webadmin with some pending patches to get this functionality. I was cautioned that the patches were fairly new, but nevertheless I was able to create VMs with varying NUMA configurations, and didn't notice any regressions. Thanks, Greg

1 0

[Mentor Opportunities] Outreach Program for Women
by Brian Proffitt 19 Sep '14

19 Sep '14

All: I am pleased to announce that oVirt will be participating in the next round of the GNOME Foundation's Outreach Program for Women, an internship program designed to foster more gender-diverse participation within the free and open source software communities. Currently, we have at least one intern position lined up for the program (with hopefully more if our board members decide to sponsor more internships). Right now, we are seeking developers who would be willing to act as mentors from the oVirt project. Details about the program and mentor requirements are available on our OPW landing page[1]. Essentially, this program is very much like the Google Summer of Code project, where we match up interns with projects within oVirt that they can finish up during this round of OPW (December 9, 2014 to March 9, 2015). There is one key difference, that mentors should note: the application process is very interactive, and potential interns are expected to work with program mentors even as they are applying. oVirt mentors will work with participants before and after the application process to help integrate potential interns into the oVirt ecosystem, and continue working with accepted interns as the program progresses. Interested mentors should do the following: 1. Check out the application process[2] for interns so you can see what they are required to do and, in turn, know what will be expected from mentors. 2. Visit the Information for Mentors section[3] of the oVirt OPW page to walk through the process. 3. Fill in the Mentor Information section[4] with your name and contact info. If you have any questions about the program, please let me know. We would like to have mentors lined up as soon as possible, given the upcoming application deadline and the need to have applicants involved with oVirt as soon as possible. We had a pretty good experience with Google Summer of Code, and it will be very cool to continue a similar experience with this new internship program! Peace, Brian -- Brian Proffitt Community Liason oVirt Open Source and Standards, Red Hat - http://community.redhat.com Phone: +1 574 383 9BKP IRC: bkp @ OFTC [1] http://www.ovirt.org/Outreach_Program_for_Women [2] http://wiki.gnome.org/OutreachProgramForWomen#Application_Process [3] http://www.ovirt.org/Outreach_Program_for_Women#Information_for_Mentors [4] http://www.ovirt.org/Outreach_Program_for_Women#Mentor_Information

1 0

[ENGINE] thoughts about LinqUtils
by Yair Zaslavsky 19 Sep '14

19 Sep '14

Hi Engine developers (and others :) ), I would like to share some thoughts & questions with you about the future of LinqUtils 1. Location of LinqUtils Currently LinqUtils resides at utils. However, as LinqUtils contains some non engine related code, I think it should be moved to the new uutils (universal utils) project. uutils was introduced during the work on AAA, in order to provide a set of utilities that are not engine related, and hopefully in the future will be published as a separate artifcat on maven central (Alon, feel free to correct me here :) ) 2. The name LinqUtils - Linq comes from c#. I think these days are way behind us (3 years or so now?) i would like to suggest renaming the class. CollectionUtils? Any other suggestions are more than welcome Thanks, Yair

11 24

[QE][ACTION REQUIRED] oVirt 3.5.0 status
by Sandro Bonazzola 19 Sep '14

19 Sep '14

Hi, We are going to start composing oVirt 3.5.0 GA (or RC3, depending on Monday morning Go / No go Meeting decisions) on Monday 2014-09-22. Maintainers: - Please be sure that 3.5 snapshot satisfy release criteria[9] - Please be sure that no pending patches are going to block the release - If any patch must block the GA release please raise the issue as soon as possible. - If any packages need a rebase please raise the issue as soon as possible. - Be aware that packages that doesn't need a rebase must be re-built with final release versioning from the RC2 tag. The bug tracker [1] shows the following proposed blockers to be reviewed: Bug ID Whiteboard Status Summary 1144079 integration POST local engine-reports-setup does not write conf file to allow accessing reports from engine 1138340 sla POST vdsm relies on unsupported 'min_guarantee' VM configuration parameter 1143873 network NEW firewall-cmd gets stuck when run as superVDSM subsubprocess 1142256 integration NEW remote engine-reports-setup does not write conf file to allow accessing reports from engine And the following dependencies still open: Bug 1041569 - [NFR] libvirt: Returning the watermark for all the images opened for writing The following bugs are keyworded as Regression and not marked as blockers[10] Bug ID Whiteboard Status Summary 1142709 infra NEW Trying to deploy hosted-engine via iSCSI device fails 1138144 storage NEW Failed to autorecover storage domain after unblocking connection with host 1118349 storage NEW [vdsm] Creating DataCenter 3.5 using master domain V1 fails with InquireNotSupportedError 1110444 ux POST bookmark selection does not work on first try 1138314 virt NEW Fail to start vm with payload. Feature freeze is now effective, and branch has been created. All new patches must be backported to 3.5 branch too. Features completed are marked in green on Features Status Table [2] There are still 86 bugs [3] targeted to 3.5.0. Excluding node and documentation bugs we still have 62 bugs [4] targeted to 3.5.0. More in detail [5]: Whiteboard NEW ASSIGNED POST Total docs 13 1 0 14 gluster 8 4 2 14 i18n 0 0 1 1 infra 0 1 2 3 integration 3 1 2 6 network 0 1 0 1 node 7 4 0 11 ppc 2 0 4 6 sla 12 0 7 19 virt 8 0 3 11 Total 53 12 21 86 Maintainers / Assignee: - Please ensure that completed features are marked in green on Features Status Table [2] - If you find a blocker bug please remember to add it to the tracker [1] - Please fill release notes, the page has been created here [6] - Please review results from Third Test Day on the etherpad [7] and on the mailing lists - Please update the target to 3.5.1 or later for bugs that won't be in 3.5.0: it will ease gathering the blocking bugs for next releases. Community: - You're welcome to join us testing last release candidate or nightly builds and getting involved in oVirt Quality Assurance[8] [1] http://bugzilla.redhat.com/1073943 [2] http://goo.gl/4SuYdE [3] http://red.ht/1pVEk7H [4] http://red.ht/1zT2mSq [5] http://red.ht/1q7SqNL [6] http://www.ovirt.org/OVirt_3.5_Release_Notes [7] http://etherpad.ovirt.org/p/3.5-testday-3 [8] http://www.ovirt.org/OVirt_Quality_Assurance [9] http://www.ovirt.org/OVirt_3.5_release-management#Release_Criteria_.28WIP.29 [10] http://goo.gl/uavikG Thanks, -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

4 3

OVIRT-3.5-TEST-DAY-3: CPU SLA for capping
by Martin Perina 19 Sep '14

19 Sep '14

Hi, I tested the support to limit CPU utilization of hosts. Setting it up in the GUI worked correctly, but for running VM settings are ignored. I used XMLRPC protocol to avoid [1]. To simulate VM CPU utilization I used stress tool [2]. Testing steps: 1. Create a VM with the same number of CPUs as Host 2. Create CPU type QoS in DC and set the limit to desired value 3. In Cluster create new CPU profile with above QoS and remove the default CPU profile 4. Start the VM and execute (N is a number of CPUs in VM) stress --cpu N 5. Connect to Host and execute sar -u 1 10 No matter what the limit was set to (1, 50, 100), host CPUs were always utilized to 100%. So I create bug to cover this [3]. Martin Perina [1] http://bugzilla.redhat.com/1142851 [2] http://people.seas.harvard.edu/~apw/stress/ [3] http://bugzilla.redhat.com/1144280

1 0

Re: [ovirt-devel] Broken dependencies: vdsm
by Balamurugan Arumugam 18 Sep '14

18 Sep '14

[Adding vdsm-devel/ovirt-devel] ----- Original Message ----- > From: "Kaleb S. KEITHLEY" <kkeithle(a)redhat.com> > To: "Balamurugan Arumugam" <barumuga(a)redhat.com>, "Dan Kenigsberg" <danken(a)redhat.com> > Cc: "Lalatendu Mohanty" <lmohanty(a)redhat.com>, buildsys(a)fedoraproject.org, "Development discussions related to > Fedora" <devel(a)lists.fedoraproject.org>, glusterfs-owner(a)fedoraproject.org > Sent: Monday, September 15, 2014 9:27:03 PM > Subject: Re: Broken dependencies: vdsm > > Posting to -devel because I can't post to vdsm-owner or virt-maintenance. > > On 09/15/2014 10:09 AM, Kaleb S. KEITHLEY wrote: > > gluster server. AFAIK it's a mistake for vdsm to have these as > > dependencies. > > Correcting myself. glusterfs-cli is okay, and should, AFAIK, be in > RHEL6, although I'm not seeing that it is. > > glusterfs-server is not in RHEL6, and won't ever be in RHEL6, and AFAIK > it's a mistake for vdsm to have it as a dependency. > Looks like to me too. Lets check with vdsm-devel/ovirt-devel? Regards, Bala

2 1

OVIRT-3.5-TEST-DAY-3: replace XML-rpc with JSON-rpc
by Francesco Romani 18 Sep '14

18 Sep '14

Hi everyone of this test day I played with the new JSON-rpc transport, which is enabled by default. I set up from scratch two hypervisors running on RHEL 6.5, while Engine was running on F19. I created everything from scratch and tested all the flows I known good enough, including less-then-common stuff. The list includes creating storage domains, VMs, hotplug/unplug disks, snapshots, migration, hibernation and dehibernation. On top of that, I prepared an environment to run more tests, that included create 200 VMs, started/stopped 100 of them to test. I kept an eye on the logs other than Engine. Everything I tried went OK, and logs look good to me. I run in a few hiccups, which I mention for the sake of completeness: - VDSM refused to start or run VMs initially: libvirt config included relics from past environment on the same box, not JSON-rpc fault. Fixed with new config and (later) a reboot. - Trying recovery, Engine took longer than expected to sync up with VDSM. I have not hard data and feeling is not enough to file a BZ, so I didn't. - Still trying recovery, one and just one time Engine had stale data from VDSM (reported two VMs as present which actually aren't). Not sure it was related to JSON-rpc, can't reproduce, so not filed a BZ. Since I need to continue profiling anyway (better late than never I guess), I include a couple of test to see if JSON-rpc makes any difference to JSON-rpc. Results will be posted on devel(a)ovirt.org and/or on the oVirt wiki. -- Francesco Romani RedHat Engineering Virtualization R & D Phone: 8261328 IRC: fromani

3 5