Hi list,
I realized I am using drbd now for more than ten years and is has been a
joy all the time.
Now I have a request where I am really unsure how to deal with.
I have to resize a volume offline. Mainly because the carppy hypervisor
does not allow to enalrge virtual disks while the vm ist
Hi all,
I have to cross-post to LVM as well to DRBD mailing list as I have no
clue where the issue is- if it's not a bug...
I can not get working LVM on top of drbd- I am getting I/O erros
followed by "diskless" state.
Steps to reproduce:
Two machine2.
A: CentOS7 x64; epel-providedd packages
Hi all,
>> But yes, any write during resync to no-yet synced, "dirty" blocks
>> will bring those blocks in sync as well.
>>
>> As an extreme example, if you'd just zero-out the full device during
>> resync, you'd bring it in-sync by application writes.
> This 'zero-out' trick is exactly what I do
Hi all,
just a simple question for my own curiosity.
Assume a two node master-slave setup. Disk on the master fails. All
reads and writes go through the slave (master remains primary).
Now the disk on the primary is replaced and the full sync is started.
Now let's say the sync is at 40% and a w
Hi,
I do not exactly know what you are talking about ;-)
So you have a drbd device which need to grow, right? On one server (or
on both?) the disk (lv) for the drbd device is a LVM logical volume,
correct?
And I guess you are talking of the PE size of the volume group which is
configured for 32M
Hi Lars again,
> look for nfs + fscache maybe.
> At least its intended use case is "almost" what you apparently want.
This looks interesting, yes.
I guess it still will slow down any writes on the remote because of
locking and data transfer. But on the main site it will keep all things
smoothly.
S
Hi Lars, and all,
> There is "geo replication" with glusterfs.
> Though that basically is also simply a masked-off rsync "cron job".
Yeah, that is it. It does not help as users are located on both sites
accessing (and editing!) the files.
> But nothing will change the fact that if you have only 10
Am 05.12.2014 um 07:54 schrieb Devin Reade:
> You might want to examine GlusterFS for that use case where users connect to
> the local server node.
>
I had a look at glusterfs. Even though the mailing list is pretty very
low traffic...
Looks like a file write will be delayed until the file is wri
Am 05.12.2014 um 07:54 schrieb Devin Reade:
> You might want to examine GlusterFS for that use case where users connect to
> the local server node.
>
Sounds pretty good, indeed. I did a very quick test and it appears to
work as expected.
Do you have some experience related to write speed if the
Hi All,
>
>>
>> Is it possible to run a two-node setup
> Yes.
>
>> *without* a clustered filesystem
> No.
Thought so. Never mind.
> But, what do you really want to solve, what exactly is the challenge?
> Lars
Well, I need to connect two sites. As money is the limiting factor we
can not have more
Hi,
>
>
>
> Run a NFS server on the master, and mount the share on the slave.
>
Well, my idea was to save bandwidth. In this setup every read is still
coming from the master. So it does not help in saving bandwidth and
additionally the local (for slave) clients do not have any performance
benefit.
Hi,
I have not ried yet and with a little bit of thinking th answer might be
obvious. But anyways:
Is it possible to run a two-node setup without a clustered filesystem
where both heads access the data but one of these two nodes only as
read-only?
I guess it is still somewhat complicated as ext4
Hi,
I have a nearly default setup running. I moved the secondary drbd to a
VM at it appears to be incredibly slow.
I need some hints to troubleshoot why it is so slow. See the cat /proc/drbd:
GIT-hash: a798fa7e274428a357657fb52f0ecf40192c1985 build by
phil@Build64R6, 2013-09-27 16:00:43
0: cs:Sy
Hi all,
I' using drbd meanwhile for some years and it's working pretty straight
forward. I really like it. Currently I'm using it just as a security
instance- writes occur only on the primary node.
Now I'm thinking of implementing it for a new setup I'll describe below:
Two sites which are conne
Hi Digimer,
>
> What you're proposing is called a "stretch cluster" and it's
> notoriously hard to do well.
>
Yeah, I realize meanwhile. As already posted I'll forget about such a
setup.
Thanks for clearance and hints how to realize anywas, but I'll better
look for other solutions.
> You said tha
Hi all,
I'm planning to use DRBD in a production environment. I prefer to use
CentOS as base system.
The reason to use drbd is the synchronisation, not the high availability.
We'll have two locations connected through a 100Mbit line. On both
locations users will access the data at the same time.
Hey,
drbd seems to work mostly really fine.
But sometimes issues occur for whatever reason. So my device went on the
secondary last night to "diskless".
Well, it didn't do any harm as the primary still was fine.
My question now is how you monitor your drbd devices? I din't have a
snmptrap devic
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi Arnold,
>
> Should be as far as I can tell from the distance. Be sure to include the
> metadata-overhead in your resize-calculations. And make sure the backup is
> fine.
Well, I did. I reduced the ext3 to 1,400GB. The drbd device to 1,450GB
and
Hi,
So I assume everything here is everything fine so far and I should able
to reduce the backing device, shouldn't I?
> Is your LVM on top of DRBD or is your DRBD on top of LVM?
>
As written- DRBD is on top of LVM. The backing device for drbd is a
logical volume. Filesystem is applied on the dr
Hi all,
as I was (lucky!) able to recover from my firt failed attempt I'll give
it now a second try.
I have drbd 8.3 running on top of LVM.
I have already reduced the filesystem (ext3) size on the drbd0 device as
well as the drbd0 device itself (I suppose I have ;-)). My last attempt
to reduce t
Hi,
I definitely performed the fsck on the drbd device, not on the lvm and
not on the physical one.
Meanwhile I was able to recover by extending the LV again to the old
size. Had to re-create the metadata on the secondary and it's syncing
now. fsck is running again, Well, weekend is safe :D
But
Hi
>> Steps I did:
>> - fsck -f (ext3)
>
> fsck of your /dev/drbdX device, I hope?
>
Yes, on the /dev/drbd0. Otherwise fsck would have thrown errors, right?
>> - reduced filesystem to 1,400G
>> - drbdadm -- --new-size=1450G resize
>> - lvreduce drbdvol -L 1500G
>> - fsck -f
>
> Again, did you
Hi,
I've a drbd device (8.3) on both sides on aLVM volume.
I tried to reduce the device now. Steps I did:
- fsck -f (ext3)
- reduced filesystem to 1,400G
- drbdadm -- --new-size=1450G resize
- lvreduce drbdvol -L 1500G
- fsck -f
- lvreduce on secondary
So far everything went fine.
After mount
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
well, I got a little bit confused by all of your answers
Soem recommended a procedure how to perform a decrease, some haven't
been aware of the state being online, but not connected...and so on.
Never mind!
In the end I just took my chance. ;
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
> ELRepo has it;
>
> rpm --import http://elrepo.org/RPM-GPG-KEY-elrepo.org rpm -Uvh
> http://elrepo.org/elrepo-release-6-4.el6.elrepo.noarch.rpm yum
> install drbd83-utils kmod-drbd83
Thanks for the links!
Syncing now for the next 10hours... :D
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
does someone of you guys know a repository for CentOS 6/64bit which
hosts drbd?
I checked rpmforge but it does not.
Any other clues?
And no, I don't want to build it on my own.
Greetings
Christian
-BEGIN PGP SIGNATURE-
Version: Gn
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
>> Otherwise I'd expect just to grow the underlying LVM and
>> then do a "drbadm resize drbd0". Will this do any harm?
yes, I'm afraid so.
>> Is there an explanation why it's not working? Even then when
>> I'll setup the upcomin
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
>> Otherwise I'd expect just to grow the underlying LVM and then do
>> a "drbadm resize drbd0". Will this do any harm?
>
> yes, I'm afraid so.
Is there an explanation why it's not working? Even then when I'll
setup the upcoming secondary from sc
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I have a two-node pri/sec setup here. Currently the secondary is
offline and I when it'll go online it'll be a fresh installed node.
Full sync then.
But I need to increase the size of the primary device now.
Having read http://www.drbd.org/users
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
> Is the new filesystem for VMWare's vSphere 5 cluster aware?
Of course it is- it's one of the major features of VMFS.
> see references to GFS2 and OCFS2 but never a mention of VMFS5.
Well, the reason might be that you won't be able to have a
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
> You can make a node primary during sync, or even when it does not
> have a backing device. Primary/secondary status has little to do
> with the status of the backing device, unless you do not have at
> least one UpToDate device in the resource.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
my primary node failed and I had to assign a new backing device to it.
During this my secondary was assigned manually as primary. Well,
worked so far.
Now as my primary is back I want to enable it again as primary of
course. Currently it is synci
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I had the primary going to diskless state so I wanted to (well,
manually this time) activate the secondary as new primary. My script
worked nice well except one think:
I tried to set the secondary as new primary with "drbdadm primary
dedb0" but i
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi David,
>> I wanted to make the secondary now the new primary with "drbdadm
>> primary drbd0" but it refused to do so because the primary was
>> still reachable by ping and network as well drbd service was
>> running.
> Did you 'drbdadm secondary d
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I have a two-node cluster here. My primary now rebooted and lost his
disk (intended, no need to worry).
Now the primary was in state "diskless", but still primary and
seconday was UptoDate.
I wanted to make the secondary now the new primary with
Hi,
> In our case, we use drbd to sync oracle db data in two node across the
> wan. Since the net bandwidth is slow(<2M), the i/o performance is
> pretty poor in primary node because i/o operation will be pending to
> wait the data transmission.
Officially wording would be to purchase the drbd pro
Hi,
I was able to solve nearly everything on my own. Of course I had to
re-configure VMware tools. No worries about this. And you're right- as
the tools where not properly configured initially drbd had no network
and no disk.
I'm just wondering why it ended up in a split-brain scenario in the
Hi,
well yes. Was wrong wording. :D
Thanks for letting me know it's supposed to not create any issues.
Greetings
Christian
___
drbd-user mailing list
drbd-user@lists.linbit.com
http://lists.linbit.com/mailman/listinfo/drbd-user
Hi all,
I had an issue here to tell- I assume this behavior is intended but this
time it caused issues even with clean shut down.
First, my setup is a single primary (Virtual Machine/ VM) and single
secondary (physical box). Due to power outage I had to shutdown the
boxes. Clear shutdown, no hard
Hi,
just a question currently:
Is it supported to have drbd running on one node as 32bit version and on
the secondary as 64bit?
We shouldn't use drbd versions but 32 vs. 64 should not be an issue?
Greetings
Christian
___
drbd-user mailing list
drbd-
Hi,
>> So how can I recover now?
>
> Consider upgrading to a recent DRBD version. For your age-old 8.2.6, the
> correct syntax is here:
>
> http://www.drbd.org/users-guide-legacy/s-resolve-split-brain.html
Great! Found it and is currently syncing.
And, no I'm not going to upgrade as this is the ve
Hi,
as I just send out I had a short outage which ended in a Split-Brain
scenario.
I'm trying to recover now from this and have all drbd devices back again.
Unfortunately I can't recover from split-brain. Could someone help me,
please?
This is the current state on the primary:
[root@backuppc ~]#
Hi Felix,
>> What? Do you say the primary updates its data from the secondary during
>> an "adjust" command?
> Generally? No. In your case? Yes :-)
And why this?
> OK, that's ancient? So that's what was shipping with RHEL5? Hmm. You may
> want to
> a) step away from your distributor's packaged DRB
Hi Felix,
> this reminds me of something I've seen during a drbdadm adjust.
> The adjusting Primary would re-initialize its DRBD and then (apparently)
> sync back its hot AL extents from the peer. Seeing as the peer was
> Inconsistent in your case (it was still syncing), this can't work, and
> the
Ok, I solved it on my own. Still don't know why this happened
I re-created the metadata with "drbdadm create-md drbd0" while the
secondary on WFConnection and then forced the primary to be primary with
"drbdsetup primary -force drbd0".
Now it's syncing and I can access my data. All good :)
C
Hi,
I'm having a two node setup here and serious issues.
node1 was running all the time without any issues.
node2 was rebooted and syncing.
As both are backed by LVM I added the "use-bmbv" in disk {} section on
both drbd.conf files. To update config I started a "drbdadm adjust
drbd0" on node 1.
Yohoo!
Why do you want to resize when you're just creating everything from scratch?
You need to be aware what you want to resize. Will the RAID volume be
enlarged at a later stage? Or do you only assign partial space from the
RAID to NFS?
Why do you want to use LVM?
If you use the full RAID arr
47 matches
Mail list logo
