similar to: [ovirt-users] Very poor GlusterFS performance

Have you tried with: performance.strict-o-direct : off performance.strict-write-ordering : off They can be changed dynamically. On 20 June 2017 at 17:21, Sahina Bose <sabose at redhat.com> wrote: > [Adding gluster-users] > > On Mon, Jun 19, 2017 at 8:16 PM, Chris Boot <bootc at bootc.net> wrote: > >> Hi folks, >> >> I have 3x servers in a

Couple of things: 1. Like Darrell suggested, you should enable stat-prefetch and increase client and server event threads to 4. # gluster volume set <VOL> performance.stat-prefetch on # gluster volume set <VOL> client.event-threads 4 # gluster volume set <VOL> server.event-threads 4 2. Also glusterfs-3.10.1 and above has a shard performance bug fix -

Dear Krutika, Sorry for asking so naively but can you tell me on what factor do you base that the client and server event-threads parameters for a volume should be set to 4? Is this metric for example based on the number of cores a GlusterFS server has? I am asking because I saw my GlusterFS volumes are set to 2 and would like to set these parameters to something meaningful for performance

On Sun, Aug 6, 2017 at 4:42 AM, Jim Kusznir <jim at palousetech.com> wrote: > Well, after a very stressful weekend, I think I have things largely > working. Turns out that most of the above issues were caused by the linux > permissions of the exports for all three volumes (they had been reset to > 600; setting them to 774 or 770 fixed many of the issues). Of course, I >

Hi all again: I'm now subscribed to gluster-users as well, so I should get any replies from that side too. At this point, I am seeing acceptable (although slower than I expect) performance much of the time, with periodic massive spikes in latency (occasionally so bad as to cause ovirt to detect a engine bad health status). Often, if I check the logs just then, I'll see those call traces

The profile seems to suggest very high latencies on the brick at ovirt1.nwfiber.com:/gluster/brick1/engine ovirt2.* shows decent numbers. Is everything OK with the brick on ovirt1? Are the bricks of engine volume on both these servers identical in terms of their config? -Krutika On Wed, May 30, 2018 at 3:07 PM, Jim Kusznir <jim at palousetech.com> wrote: > Hi: > > Thank you. I

I've been back at it, and still am unable to get more than one of my physical nodes to come online in ovirt, nor am I able to get more than the two gluster volumes (storage domains) to show online within ovirt. In Storage -> Volumes, they all show offline (many with one brick down, which is correct: I have one server off) However, in Storage -> domains, they all show down (although

[Adding gluster-users back] Nothing amiss with volume info and status. Can you check the agent.log and broker.log - will be under /var/log/ovirt-hosted-engine-ha/ Also the gluster client logs - under /var/log/glusterfs/rhev-data-center-mnt-glusterSD<volume>.log On Wed, May 30, 2018 at 12:08 PM, Jim Kusznir <jim at palousetech.com> wrote: > I believe the gluster data store for

Adding Ravi to look into the heal issue. As for the fsync hang and subsequent IO errors, it seems a lot like https://bugzilla.redhat.com/show_bug.cgi?id=1497156 and Paolo Bonzini from qemu had pointed out that this would be fixed by the following commit: commit e72c9a2a67a6400c8ef3d01d4c461dbbbfa0e1f0 Author: Paolo Bonzini <pbonzini at redhat.com> Date: Wed Jun 21 16:35:46 2017

https://bugzilla.netfilter.org/show_bug.cgi?id=871 Summary: Running two instances of ulog causes abort in libnfnetlink Product: ulogd Version: SVN (please provide timestamp) Platform: All OS/Version: All Status: NEW Severity: critical Priority: P5 Component: ulogd AssignedTo:

On 16/10/13 06:40, David N Melik wrote: > On Tue, 15 Oct 2013 21:56:16 -0400 > Charles Lepple <clepple at gmail.com> wrote: >>>> > >> What are you trying to optimize? >>> > > >>> > > nothing, but I just did not want another daemon running, or at >>> > > least I do not want a port open that can access the UPS, because I

On Thu, May 31, 2018 at 3:16 AM, Jim Kusznir <jim at palousetech.com> wrote: > I've been back at it, and still am unable to get more than one of my > physical nodes to come online in ovirt, nor am I able to get more than the > two gluster volumes (storage domains) to show online within ovirt. > > In Storage -> Volumes, they all show offline (many with one brick down,

On Wed, 16 Oct 2013 08:19:11 +0100 Chris Boot <bootc at bootc.net> wrote: > >> > I forgot to mention: by default, NUT listens on localhost. If > >> > you are using Linux, you could add a "-m owner --uid-owner" rule > >> > to iptables to only match the UID for the NUT system user. > > > > Thanks; if it listens on localhost by default,

On 19/09/13 01:30, Kris Jordan wrote: > Luke-Jr wrote, On 9/18/2013 1:02 PM: >> I wanted to suggest some rating for functionality. I got the UPS >> available >> that seemed to have the best possible rating from NUT's list (Eaton >> Powerware >> UPS 1500), and found out it doesn't support telling me battery level or >> runtime! :/ > > I'm

No, you don't need to do any of that. Just executing volume-set commands is sufficient for the changes to take effect. -Krutika On Wed, Jun 21, 2017 at 3:48 PM, Chris Boot <bootc at bootc.net> wrote: > [replying to lists this time] > > On 20/06/17 11:23, Krutika Dhananjay wrote: > > Couple of things: > > > > 1. Like Darrell suggested, you should enable

On Tue, 15 Oct 2013 21:56:16 -0400 Charles Lepple <clepple at gmail.com> wrote: > >> What are you trying to optimize? > > > > nothing, but I just did not want another daemon running, or at > > least I do not want a port open that can access the UPS, because I > > will not be using it; it is more of a minor security issue than > > anything. The

[replying to lists this time] On 20/06/17 11:23, Krutika Dhananjay wrote: > Couple of things: > > 1. Like Darrell suggested, you should enable stat-prefetch and increase > client and server event threads to 4. > # gluster volume set <VOL> performance.stat-prefetch on > # gluster volume set <VOL> client.event-threads 4 > # gluster volume set <VOL>

Luke-Jr wrote, On 9/18/2013 1:02 PM: > I wanted to suggest some rating for functionality. I got the UPS available > that seemed to have the best possible rating from NUT's list (Eaton Powerware > UPS 1500), and found out it doesn't support telling me battery level or > runtime! :/ I'm avoiding Eaton/Powerware because the lack of usbhid support, the bcmxcp driver

I wanted to suggest some rating for functionality. I got the UPS available that seemed to have the best possible rating from NUT's list (Eaton Powerware UPS 1500), and found out it doesn't support telling me battery level or runtime! :/ Luke

Hi all, I'm writing the firmware for a home-made 12V UPS, which will be used to power low-voltage equipment in my home like my ADSL modem, router, switches and a Soekris net4801. The home-made UPS can run those devices on the same batteries as my 230V UPS, but last for several hours rather than a few minutes due to increased efficiency. The microcontroller I'm using has a serial