Ceph at the time, I found it that it
required more moving parts (compared to Gluster which is of course silly
simple to get going) for not much improvement, but i'd assume they have
improved that since then.
-wk
Kind regards,
Alex.
/Z
On Sat, 28 Oct 2023 at 11:21, Strah
to be refreshed, it had been up for
hundreds of days.
-wk
On 7/21/23 12:02 PM, W Kern wrote:
we have an older 2+1 arbiter gluster cluster running 6.10 on Ubuntu18LTS
It has run beautifully for years. Only occaisionally needing attention
as drives have died, etc
Each peer has two volumes. G1 and G2
normal ops and even when healing after host
maintenance its less than 100Mb/s
-wk
On 12/31/2022 12:43 AM, Alan Orth wrote:
Hi Filipe,
I think it would probably be fine. The Red Hat Storage docs list the
important thing being /5ms latency/, not link speed:
https://access.redhat.com
OK, so it seems 8.x is the way to go
So what about Thin Arbiter?
Is anyone one using it in production?
-wk
On 12/18/2020 12:59 AM, Olaf Buitelaar wrote:
It is in their release notes;
https://docs.gluster.org/en/latest/release-notes/7.9/
<https://docs.gluster.org/en/latest/release-no
OK, so it seems 8.x is the way to go
So what about Thin Arbiter?
Is anyone one using it in production?
-wk
On 12/18/2020 12:59 AM, Olaf Buitelaar wrote:
It is in their release notes;
https://docs.gluster.org/en/latest/release-notes/7.9/
<https://docs.gluster.org/en/latest/release-no
On 12/17/2020 10:13 AM, Olaf Buitelaar wrote:
Hi WK,
i believe gluster 7 just received it's
latest maintenance patch @version 7.9. so that's a dead end from there on.
Hmm, ok, the website should reflect that then.
I'm not sure if you can skip a whole version going straig
biter?
Is it considered Safe?
How much of an improvement of people seen with it?
For Us, Classic Arbiter works great, so we would need a reaon to go with
Thin, but speed would be one of them if we are giving up safety.
-wk
Community Meeting Calendar:
Schedule -
Every 2nd and 4th
d two or more Gbit/s
ports. Round-Robin teamd is really easy to setup, or use the traditional
bonding in its various flavors. You probably have some spare NIC cards
lying around so its usually a 'freebie'
Of course best case would be to make the jump to 10Gb/s kit.
-wk
sorry, I didn't notice you had already looked at gstatus.
Nonetheless with its JSON output you certainly cover the issues you
described
i.e. "When Brick went down (crash, failure, shutdown), node failure,
peering issue, on-going healing"
which is how we use it.
-wk
On 10/
https://github.com/gluster/gstatus
we run this from an ansible driven cronjob and check for the healthy
signal in status, as well as looking for healing files that seem to persist.
We have a number of gluster clusters and we have found its warnings both
useful and timely.
-wk
On 10/26
-type none
Finally I used Karthik's remove-brick command and it worked this time
and I am now copying off the needed image.
So I guess order counts.
Thanks.
-wk
On 8/27/2020 12:47 PM, WK wrote:
No Luck. Same problem.
I stopped the volume.
I ran the remove-brick command. It warned about
No Luck. Same problem.
I stopped the volume.
I ran the remove-brick command. It warned about not being able to
migrate files from removed bricks and asked if I want to continue.
when I say 'yes'
Gluster responds with 'failed: Quorum not met Volume operation not allowed'
set commands due
to the quorum and spits out the set failed error.
So in modern Gluster, what is the preferred method for starting and
mounting a single node/volume that was once part of a actual 3 node
cluster?
Thanks.
-wk
Community Meeting Calendar:
Schedule -
Every 2nd and
but didn't really
notice much of a difference. You get a smaller number of heals but they
are bigger and take longer to sync.
Does anyone know why the difference and the reasoning involved?
-WK
Community Meeting Calendar:
Schedule -
Every 2nd and 4th Tuesday at 14:30
On 12/12/2019 4:34 AM, Ravishankar N wrote:
On 12/12/19 4:01 am, WK wrote:
so I can get some sort of resolution on the issue (i.e. is it
hardware, Gluster etc)
I guess what I really need to know is
1) Node 2 complains that it cant reach node 1 and node 3. If this
was an OS/Hardware
node3 have error message complaining about not reaching
node2
2) how significant is it that the node was running 6.5 while node 1 and
node 2 were running 6.4
-wk
Forwarded Message
Subject:VM freeze issue on simple gluster setup.
Date: Thu, 5 Dec 2019 16:23:35 -0800
we have a large of these replica 2+1 gluster setups running
gluster version from 5.x up and none of the others have had this issue
Any advise would be appreciated.
Sincerely,
Wk
Community Meeting Calendar:
APAC Schedule -
Every 2nd and 4th Tuesday at 11:30 AM IST
Bridge: http
great, I'll dig into when I get a chance and report back.
-wk
On 10/14/2019 10:05 AM, Amar Tumballi wrote:
Awesome, thanks!
Then, I hope this
https://github.com/gluster/glusterfs/commit/ab2558a3e7a1b2de2d63a3812ab4ed58d10d8619 is
included in the build.
What it means is, if you just
r VM file.
We eventually figured out the problem VM using other methods and
resolved the issue, but we would still like to know if there is a
script
or recipe to determine what file a shard may belong to as that would
have sped up the resolution.
-wk
Hi WK,
There are a few patches [1] that are still undergoing review . It
would be good to wait for some more time until trying it out. If you
are interested in testing, I'll be happy to inform you once they get
merged.
[1] https://review.gluster.org/#/c/20095/,
On 4/23/2018 11:46 AM, Jamie Lawrence wrote:
Thanks for that, WK.
Do you know if those images were sparse files? My understanding is that this
will not work with files with holes.
We typically use qcow2 images (compat 1.1) with metadata preallocated
(so yes, sparse) So we may be
hing
up. In the happy case, you can test it by comparing the md5sum of the
file from the mount to that of your stitched file."
We tested it with some VM files and it indeed worked fine. That was
probably on 3.10.1 at the time.
-wk
On 4/20/2018 12:44 PM, Jamie Lawrence wrote:
Hel
+1
I also would like to see those instruction.
I've been interested in NFS-Ganesha with Gluster but there wasn't
obvious references that were up to date.
(Let alone introduction of StorHaug)
-wk
On 2/25/2018 10:28 PM, Serkan Çoban wrote:
I would like to see the steps for refe
I'm reading that the new kernel will slow down context switches. That is
of course a big deal with FUSE mounts.
Has anybody installed the new kernels yet and observed any performance
degradation?
-wk
___
Gluster-users mailing list
Gluster-
I definately prefer replica 2 arbiter 1.
It makes more sense and is more accurate, since that scenario has only
two copies of the actual data.
-wk
On 12/20/2017 2:14 AM, Ravishankar N wrote:
Hi,
The existing syntax in the gluster CLI for creating arbiter volumes is
`gluster volume
Will Do, re post as an issue.
I may repost on Monday about if the virt settings I am using are current
for 3.12,
I've also seen that this list is dead on the weekend.
-wk
On 10/20/2017 11:25 AM, Kaleb S. KEITHLEY wrote:
On 10/20/2017 01:06 PM, WK wrote:
2. Can someone get the co
newbie to be able to get a hold of easily.
-wk
___
Gluster-users mailing list
Gluster-users@gluster.org
http://lists.gluster.org/mailman/listinfo/gluster-users
rse most of those reviews were when there was a significant
price difference.
-wk
___
Gluster-users mailing list
Gluster-users@gluster.org
http://lists.gluster.org/mailman/listinfo/gluster-users
You have replica2 so you can't really take 50% of your cluster down
without turning off quorum (and risking split brain). So detaching the
rebuilding peer is really not an option.
If you had replica3 or an arbiter, you CAN detach or isolate the problem
peer. I've done things like change the G
Just so I know.
Is it correct to assume that this corruption issue is ONLY involved if
you are doing rebalancing with sharding enabled.
So if I am not doing rebalancing I should be fine?
-bill
On 10/3/2017 10:30 PM, Krutika Dhananjay wrote:
On Wed, Oct 4, 2017 at 10:51 AM, Nithya Balach
nd arbiter improvements.
I therefore don't have any baseline stats to compare any performance diffs.
I'm curious as to what changed in 3.10 that would account for any change
in performance from 3.8 and in a similar vein what changes to expect in
3.12.x as we are thinking about making t
On 9/10/2017 2:02 AM, Pavel Szalbot wrote:
WK: I use bonded 2x10Gbps and I do get crashes only in heavy I/O
situations (fio). Upgrading system (apt-get dist-upgrade) was ok, so
this might be even related to amount of IOPS.
-ps
Well, 20Gbps of writes could overwhelm a lot of DFS clusters
;s been working perfectly
fine. And trust me, there had been A LOT of various crashes, reboots and
kill of nodes.
Maybe it's a version thing ? A new bug in the new gluster releases that
doesn't affect our 3.7.15.
On Sat, Sep 09, 2017 at 10:19:24AM -0700, WK wrote:
Well, that makes m
On 9/8/2017 11:05 PM, Pavel Szalbot wrote:
When we return the c1g node, we do see a "pause" in the VMs as the shards
heal. By pause meaning a terminal session gets spongy, but that passes
pretty quickly.
Hmm, do you see any errors in VM's dmesg? Or any other reasons for "sponginess"?
No, i
ns out to
be a potential explanation
-wk
On 9/9/2017 6:49 AM, Pavel Szalbot wrote:
Yes, this is my observation so far.
On Sep 9, 2017 13:32, "Gionatan Danti" <mailto:g.da...@assyoma.it>> wrote:
So, to recap:
- with gfapi, your VMs crashes/mount read-only with a s
I've always wondered what the scenario for these situations are (aside
from the doc description of nodes coming up and down).
Aren't Gluster writes atomic for all nodes? I seem to recall Jeff Darcy
stating that years ago.
So a clean shutdown for maintenance shouldn't be a problem at all. If
e using oVirt/RHEV. Is it possible that your platform is
triggering a protective response on the VMs (by suspending).
-wk
On 9/8/2017 5:13 AM, Gandalf Corvotempesta wrote:
2017-09-08 14:11 GMT+02:00 Pavel Szalbot :
Gandalf, SIGKILL (killall -9 glusterfsd) did not stop I/O after few
minutes.
On 8/25/2017 2:21 PM, lemonni...@ulrar.net wrote:
This concern me, and it is the reason I would like to avoid sharding.
How can I recover from such a situation? How can I "decide" which
(reconstructed) file is the one to keep rather than to delete?
No need, on a replica 3 that just doesn't happ
On 8/25/2017 12:43 PM, lemonni...@ulrar.net wrote:
I think you are talking about DRBD 8, which is indeed very easy. DRBD 9
on the other hand, which is the one that compares to gluster (more or
less), is a whole other story. Never managed to make it work correctly
either
Yes, and I noticed
On 8/25/2017 12:56 AM, Gionatan Danti wrote:
WK wrote:
2 node plus Arbiter. You NEED the arbiter or a third node. Do NOT try 2
node with a VM
This is true even if I manage locking at application level (via
virlock or sanlock)?
We ran Rep2 for years on 3.4. It does work if you are
On 8/23/2017 10:44 PM, Pavel Szalbot wrote:
Hi,
On Thu, Aug 24, 2017 at 2:13 AM, WK wrote:
The default timeout for most OS versions is 30 seconds and the Gluster
timeout is 42, so yes you can trigger an RO event.
I get read-only mount within approximately 2 seconds after failed IO.
Hmm
That really isnt an arbiter issue or for that matter a Gluster issue. We
have seen that with vanilla NAS servers that had some issue or another.
Arbiter simply makes it less likely to be an issue than replica 2 but in
turn arbiter is less 'safe' than replica 3.
However, in regards to Gluster
s the source.
We are on 3.10 now and the arbiter+sharding go a long way in solving
that issue.
-wk
___
Gluster-users mailing list
Gluster-users@gluster.org
http://lists.gluster.org/mailman/listinfo/gluster-users
On 8/6/2017 1:09 PM, lemonni...@ulrar.net wrote:
Are your gluster nodes physically isolated on their own network/switch?
Nope, impossible to do for us
ok, yes, that makes it much harder to secure.
You should add VLANS, and/or overlay networks and/or Mac Address
filtering/locking/security
I'm not sure what you mean by saying "NFS is available by anyone"?
Are your gluster nodes physically isolated on their own network/switch?
In other words can an outsider access them directly without having to
compromise a NFS client machine first?
-bill
On 8/6/2017 7:57 AM, lemonni...@ulrar
ds
or so while it figures things out and then continue on.
When the missing node returns, the self-heal will kick in and you will
be back to 100%.
Your other alternative is to turn off quorum. But that risks
split-brain. Depending upon your data, that may or may not be a seriou
I finally did find some stats on teaming
http://rhelblog.redhat.com/2014/06/23/team-driver/
On 6/19/2017 10:42 AM, WK wrote:
OK, at least its not an *issue* with Gluster. I didn't expect any but
you never know.
I have been amused at the 'lack' of discussion on Teaming per
ed any bugs or issues
using teaming.
*/David Gossage/*/*
*/
//*Carousel Checks Inc.| System Administrator*
*Office*708.613.2284
On Sat, Jun 17, 2017 at 2:59 PM, wk <mailto:wkm...@bneit.com>> wrote:
I'm looking at tuning up a new site and the bonding issue came up
I'm looking at tuning up a new site and the bonding issue came up
A google search reveals that the gluster docs (and Lindsay) recommend
balance-alb bonding.
However, "team"ing came up which I wasn't familiar with. Its already in
RH6/7 and Ubuntu and their Github page implies its stable.
The
On 6/10/2017 5:12 PM, Lindsay Mathieson wrote:
Three good nodes - vnb, vng, vnh and one dead - vna
from node vng:
root@vng:~# gluster peer status
Number of Peers: 3
Hostname: vna.proxmox.softlog
Uuid: de673495-8cb2-4328-ba00-0419357c03d7
State: Peer in Cluster (Disconnected)
Hostname: vn
On 6/10/2017 4:38 PM, Lindsay Mathieson wrote:
Since my node died on friday I have a dead peer (vna) that needs to be
removed.
I had major issues this morning that I haven't resolve yet with all
VM's going offline when I rebooted a node which I *hope * was due to
quorum issues as I now hav
On 5/30/2017 4:19 PM, Ric Wheeler wrote:
On 05/30/2017 06:54 PM, WK wrote:
Why is RedHat not interested in Gluster in OpenStack?
Its obvious from my years lurking on the Openstack Mailing lists that
the OpenStack community is most comfortable with Ceph. When asked
about Gluster, I
On 5/30/2017 3:24 PM, Ric Wheeler wrote:
As a community, each member needs to make sure that their specific use
case has the resources it needs to flourish. If some team cares about
Gluster in openstack, they should step forward and provide the
engineering and hardware resources needed to m
On 5/28/2017 9:24 PM, Ravishankar N wrote:
I think you should try to find if there were self-heals pending to
gluster1 before you brought gluster2 down or the VMs should not have
paused.
yes, if I watch for and then force outstanding heals (if the self-heal
hasn't kicked in) prior to shutt
On 5/28/2017 9:24 PM, Ravishankar N wrote:
Just to elaborate further, if all nodes were up to begin with and
there were zero self-heals pending, and you only brought down only
gluster2, writes must still be allowed. I guess in your case, there
must be some pending heals from gluster2 to glust
On 5/21/2017 7:00 PM, Ravishankar N wrote:
On 05/22/2017 03:11 AM, W Kern wrote:
gluster volume set VOL cluster.quorum-type none
from the remaining 'working' node1 and it simply responds with
"volume set: failed: Quorum not met. Volume operation not allowed"
how do you FORCE gluster to ign
Hah, we the opposite problem. We have a resident FireWall Nazi, who is
involved when boxes go online.
You practically have to have a court order to be able to open up a port
on any box/instance, no matter how restricted
"No port for you!"
-wk
On 1/19/2017 2:30 PM, Kevin Lemon
older Gluster boxes, still on 3.4.
NFS is turned off, perhaps deliberately by the tech who installed it.
On the newer versions, I was under the impression that NFS was
deprecated in favor of Ganesha?
Is that turned on by default?
-wk
On 1/19/2017 1:02 AM, Kevin Lemonnier wrote:
In a pinch you
for a storage
migration, and avoid the NFS setup.
You just make sure the mount points are identical as well as defined as
a POOL and of course since you don't get any locking, you also make sure
only one thing occurs at a time.
but then you use "libvirt b
Prices seem to be dropping online at NewEgg etc and going from 2 nodes
to 3 nodes for a quorum implies a lot more traffic than would be
comfortable with 1G.
Any NIC/Switch recommendations for RH/Cent 7.x and Ubuntu 16?
-wk
___
Gluster-users
ugh never mind,
I see the other thread.
On 6/29/2016 11:55 AM, WK wrote:
glad you are back up but what was the issue?
was it a function of the upgrade or something wrong with 3.7.12
-wk
On 6/29/2016 7:35 AM, Lindsay Mathieson wrote:
On 29/06/2016 11:27 PM, Kaushal M wrote:
You need to
glad you are back up but what was the issue?
was it a function of the upgrade or something wrong with 3.7.12
-wk
On 6/29/2016 7:35 AM, Lindsay Mathieson wrote:
On 29/06/2016 11:27 PM, Kaushal M wrote:
You need to edit the `/var/lib/glusterd/glusterd.info` manually on all
the nodes to reduce
using the upcoming 3.7.12
release or go straight to the upcoming 3.8.x branch (and avoid a rip and
replace upgrade later on).
-wk
On 6/22/2016 4:18 PM, Lindsay Mathieson wrote:
Hey all, have been away for two weeks and I see there has been a 3.8
release with some fascinating new features and
ve not yet upgraded to 3.7.x yet (still on 3.4 cuz it aint broke)
and are hoping that sharding solves that problem. But it seems
everytime it looks like things are 'safe' for 3.7.x, something comes up.
Fortunately, we like the fuse mount so maybe we are still ok.
-wk
On 5/16/2016 4:42 A
64 matches
Mail list logo