[ovirt-users] Re: Libgfapi considerations

2021-07-01 Thread Alexandr Mikhailov
Hi! I have tested also with last version of Ovirt 4.4, and I see that writing speed in 4 times more then with fuse. I think this is important reason to support snapshots with Libgfapi. Regards, Alex ___ Users mailing list -- users@ovirt.org To

[ovirt-users] Re: Libgfapi considerations

2021-07-01 Thread Alexandr Mikhailov
Hi! I have tested also with last version of Ovirt 4.4, and I see that writing speed in 4 times more then with fuse. I think this is important reason to support snapshots with Libgfapi. Regards, Alex ___ Users mailing list -- users@ovirt.org To unsubsc

[ovirt-users] Re: Libgfapi considerations

2021-03-29 Thread adrianquintero
Hi all, I can confirm that when using libgfapi with oVirt + Gluster replica 3 (Hyperconverged) read and write performance under a VM was 4 to 5 times better than when using fuse. -- Tested with a VM

[ovirt-users] Re: Libgfapi considerations

2021-02-11 Thread Guillaume Pavese
Hi Ritesh Chitawar, Those bugs mostly all depended on a old (2011!) qemu bug that took a very long time to be resolved : https://bugzilla.redhat.com/show_bug.cgi?id=760547 However in the meantime, the oVirt bugs that you spoke about were closed/deferred for reasons like "no activity on blocking bu

[ovirt-users] Re: Libgfapi considerations

2021-02-11 Thread Ritesh Chikatwar
Hello, There are many issues after enabling libgfapi . Issues like Attempting to start VM,also i think making a snapshot of VM, live storage migration(As Strahil Nikolov mentioned ) and VM's are not highly available etc. Because of this reason it is not enabled by default in Ovirt. Regards Rites

[ovirt-users] Re: Libgfapi considerations

2021-02-10 Thread Guillaume Pavese
Additionally to posting benchmark results and interests in the Bugzilla entries mentioned previously, I could be useful to also post in this one that should act as a RFE : [gfapi] Support libgfapi access to the gluster storage domains https://bugzilla.redhat.com/show_bug.cgi?id=1633642 Guillaum

[ovirt-users] Re: Libgfapi considerations

2021-02-10 Thread Guillaume Pavese
I strongly invite you to post those results in RedHat's Buzilla entries Guillaume Pavese Ingénieur Système et Réseau Interactiv-Group On Wed, Feb 10, 2021 at 6:55 PM wrote: > Hey everyone, > > Couple of months ago i benchmarked FUSE, libgfapi performance. If read > speed is more or less tolera

[ovirt-users] Re: Libgfapi considerations

2021-02-10 Thread scroodj
Hey everyone, Couple of months ago i benchmarked FUSE, libgfapi performance. If read speed is more or less tolerable for both, but write on FUSE is a disaster. Here below is a result screen: https://ibb.co/vBVB0WY BR Aleksandr > I recently learned that gluster community is archiving the libgf

[ovirt-users] Re: Libgfapi considerations

2021-02-05 Thread Strahil Nikolov via Users
I recently learned that gluster community is archiving the libgfapi stuff.I think that a lot of effort was spent on FUSE to get it faster.When did anyone compare them ? Best Regards,Strahil Nikolov Sent from Yahoo Mail on Android On Fri, Feb 5, 2021 at 13:12, Guillaume Pavese wrote: __

[ovirt-users] Re: Libgfapi considerations

2021-02-05 Thread Guillaume Pavese
Hey Jayme, everyone I saw that most related bugs are closed wontfix and a comment that not enough performance increase was found. I think it would help if you could update the related bugzilla entries with the performance results that you observed. Anyone interested in this feature, or with bench

[ovirt-users] Re: Libgfapi considerations

2019-12-18 Thread Jayme
It would be nice to see some progress. I have no idea why there wouldn’t be interest in adding to rhev. The io performance increase I saw while testing was phenomenal On Wed, Dec 18, 2019 at 9:42 PM Guillaume Pavese < guillaume.pav...@interactiv-group.com> wrote: > The bug has been closed wontfix

[ovirt-users] Re: Libgfapi considerations

2019-12-18 Thread Guillaume Pavese
The bug has been closed wontfix for a lack of perceived progress on the issue : https://bugzilla.redhat.com/show_bug.cgi?id=1633642 https://bugzilla.redhat.com/show_bug.cgi?id=1484227 However when following the related opened bugs in qemu, i get the feeling things are getting ready to have libgfap

[ovirt-users] Re: Libgfapi considerations

2019-12-16 Thread Jayme
I believe the snapshot issue is only present with gluster replica 3 volumes. I can confirm it on my replica 3 cluster On Mon, Dec 16, 2019 at 4:18 PM Alex McWhirter wrote: > I also use libgfapi in prod. > > > 1. This is a pretty annoying issue, i wish engine-config would look to see > if it alr

[ovirt-users] Re: Libgfapi considerations

2019-12-16 Thread Alex McWhirter
I also use libgfapi in prod. 1. This is a pretty annoying issue, i wish engine-config would look to see if it already enabled and just keep it that way. 2. Edit /etc/libvirt/qemu.conf and set dynamic ownership to 0, will stop the permission changes. 3. I don't see this error on any of my cl

[ovirt-users] Re: Libgfapi considerations

2019-12-16 Thread Jayme
The performance is certainly attractive from the minimal testing I've done with it (almost 5x I/O performance increase). For my environment I'm hitting the snapshot bug on replica 3 setups so I cannot snapshot VMs and doing so breaks the VM. This is a deal breaker for me since the VM backup softw

[ovirt-users] Re: Libgfapi considerations

2019-12-16 Thread Darrell Budic
I use libgfap in production, the performance is worth a couple of quirks for me. - watch major version updates, they’ll silently turn it off because the engine starts using a new version variable - VM/qemu security quirk that resets ownership when the VM quits, was supposedly fixed in 4.3.6 but

[ovirt-users] Re: Libgfapi considerations

2019-12-14 Thread Strahil Nikolov
According to GlusterFS Storage Domain the feature is not the default as it is incompatible with Live Storage Migration. Best Regards,Strahil Nikolov В събота, 14 декември 2019 г., 17:06:32 ч. Гринуич+2, Jayme написа: Are there currently any known issues with using libgfapi in the lat