Re: [DRBD-user] parted to move a partition DRBD uses as a backing device

On 07/31/2014 03:01 PM, Digimer wrote: > Can I assume that you don't think DRBD will care about absolute positions? Yes. That would be extra weird. Cheers, Felix ___ drbd-user mailing list drbd-user@lists.linbit.com http://lists.linbit.com/mailman/listi

Re: [DRBD-user] parted to move a partition DRBD uses as a backing device

On 07/31/2014 12:49 AM, Digimer wrote: > > However, it lists in the docs that the "FS" needs to support this, and > well, DRBD is not an FS. Weird. FWIW, I read that as "do this only if the data on the partition cares not about absolute positions of its blocks on the device", i.e. relative add

Re: [DRBD-user] Command 'drbdmeta 0 v08 /dev/sda3 internal create-md' terminated with exit code 40

On 07/22/2014 09:47 PM, Colom, Jaime wrote: > # mkfs.ext4 /dev/sda3 > > # drbdadm create-md clusterdb_res You are creating a filesystem on the backing device. There is literally no reason to do that. Defer that step to after DRBD initialization. DRBD even spells out your options: On 07/22/2014

Re: [DRBD-user] proto c - corrupt files - directories missing

Hi, wouldn't a simple disconnect/reconnect do the trick? After the handshake, drbd will want to sync all blocks with non-zero bits in the bitmap. HTH, Felix On 01/07/2014 04:51 PM, Christian Hammers wrote: > Try "drbdadm invalidate clusterdb_res" on your *secondary* node. > This will start a co

Re: [DRBD-user] drbd and jffs2(or ubifs)

Wait, has this been unanswered for two months? Wow. DRBD is generally file system agnostic, as long as you stay in the two-node active-passive paradigm. Make sure the node on which you try and create the FS is primary at the time. Take a look at the User's Guide for more details. HTH, Felix On

Re: [DRBD-user] Integrity

Hi, On 10/28/2013 10:20 PM, Pol Hallen wrote: > For purpose test on node2 I done: > > if=/dev/zero of=/dev/sdb1 count=1 bs=1M don't. There can be only pain in meddling with your backing devices. > When drbd discovers that differences? Never, unless you run a verify. When you "accidentally

Re: [DRBD-user] Ask about drbd 3 way replicate.

Hi, On 10/14/2013 03:57 AM, Watthanachai Kekhua wrote: > Dear DRBD-dev,If I clear meta data on internal disk , It that's not me but... > will destroy data also , Do you have recommend way to implement 3 way > replication on my case ? > > Found ext3 filesystem > > 19164920 k

Re: [DRBD-user] performance issue when writing to drbd devices

Hi, On 09/26/2013 05:23 PM, Matthias Teege wrote: > - no fancy drbd tunables set (e.g. "rate") heh, "fancy"? If you didn't touch any tuning parameters, low performance isn't all that miraculous. > - local logical volume -> local logical volume: FAST writing speeds > - local LVM2 logical volume -

Re: [DRBD-user] Failed action not configured

Hi, On 09/26/2013 09:35 AM, Antonio Fernández Pérez wrote: > I have configured two nodes using DRBD+Pacemaker+Corosync+MySQL: Node1 > with ip address 192.168.1.38, Node2 with ip address 192.168.1.39. > Virtual IP is 192.168.1.37. > > First problem: When I type ifconfig on Node1 only appears one I

Re: [DRBD-user] DRBD 8.3.15 on RHEL5

Hi, if your RedHat version ships RHCS, there may be no pacemaker support. So you might have to try and set it all up manually. That appears Very Hard, though. Is an OS upgrade or change to, say, Debian a possibility? Failing that, you may indeed want to try RHCS once more. HTH, Felix On 09/25/2

Re: [DRBD-user] Primary / Unknown

Hmm, in other words, you are purposefully putting your nodes in a split brain situation? It's a good thing you think of this case. Because by the nature of a Primary/Primary setup, there can be no (safe) automatic split brain recovery. As soon as the link drops, the datasets on your notes can devi

Re: [DRBD-user] drbd+mysql+innodb

On 06/13/2013 04:23 PM, Robinson, Eric wrote: > 500 instances of MySQL running for 7 years works out to 3500 years of MySQL > run-time. That's a pretty fair amount of luck. Maybe MyISAM is not as bad as > some people suggest? I never implied that it was. It just doesn't really mix with drbd. Mo

Re: [DRBD-user] drbd+mysql+innodb

Hi, On 06/13/2013 02:59 PM, Robinson, Eric wrote: > Thanks for the good input, especially from Paul and Dirk. It may help > people to understand that we have approximately 500 instances of MySQL > distributed across a dozen servers in 6 pacemaker+drbd clusters. Each > cluster consists of 2 nodes,

Re: [DRBD-user] drbd + pacemaker + diskless

Hi, has this been answered? Basically, you could implement health metrics in newer Pacemaker versions. However, I'd advise to not let diskless state get in the way of failover decisions and implement monitoring for this instead. This way, you can initiate manual failovers if appropriate. HTH, Fe

Re: [DRBD-user] How can I expand an LVM VG after growing the drbd disk

Hi, I suspect that your DRBDs act as PVs that already belong to the VGs? Then it will probably take naught but pvresize /dev/drbdX HTH, Felix On 04/26/2013 05:00 PM, Maurits van de Lande wrote: > Hello, > > > > I have recently expanded my drbd disks. I would like to add the extra > space to

Re: [DRBD-user] BUG: Uncatchable DRBD out-of-sync issue

On 04/18/2013 12:20 PM, Stanislav German-Evtushenko wrote: >> Note that your kernel (and hence kvm/virtio) can be considered rather old by >> now. > This is a stable RHEL 6 kernel at the moment. Exactly ;-) Same for Debian 6, which I no longer consider fit for KVM setups (without backports and s

Re: [DRBD-user] BUG: Uncatchable DRBD out-of-sync issue

On 04/18/2013 08:26 AM, Stanislav German-Evtushenko wrote: > If I change VIRTIO to IDE then no issues. Fascinating. Thanks for sharing! Note that your kernel (and hence kvm/virtio) can be considered rather old by now. You may see better mileage with the more recent longterm kernels such as 3.4 or

Re: [DRBD-user] Not able to test Automatic split brain recovery policies

On 04/11/2013 10:53 PM, Felix Frank wrote: > chiming in on Dan's final comment ...and then I saw Digimer's. D'oh. ___ drbd-user mailing list drbd-user@lists.linbit.com http://lists.linbit.com/mailman/listinfo/drbd-user

Re: [DRBD-user] Not able to test Automatic split brain recovery policies

Hi, On 04/10/2013 07:57 AM, Shailesh Vaidya wrote: > after-sb-1pri discard-secondary; chiming in on Dan's final comment: I hope you are aware that this setting can potentially destroy (lots of) valuable data, if for any reason the nodes reconnect and the one with the more recent dataset happens t

Re: [DRBD-user] Problem with Failover with DRBD, Heartbeat, and Pacemaker

On 04/03/2013 12:15 PM, Prater, James K. wrote: > His configuration is not failing over. Duh, you're right of course! Serves me right for posting before coffee ;-) ___ drbd-user mailing list drbd-user@lists.linbit.com http://lists.linbit.com/mailman/lis

Re: [DRBD-user] Problem with Failover with DRBD, Heartbeat, and Pacemaker

Hi, On 04/02/2013 09:31 PM, Jared Harvey wrote: > The problem arises when I put the primary node to standby. The same > thing happens when I reboot, power off, or pull the plug. it actually works as designed. > /Master/Slave Set: ms_g_drbd/ > > / Slaves: [ hpsata02 ]/ > > / Stopped: [

Re: [DRBD-user] Uncatchable DRBD out-of-sync issue

On 03/25/2013 03:50 PM, Dan Barker wrote: > You don't need Dual-Primary to live migrate - you need shared storage. Two > completely different concepts. Now, the shared storage can be based on a DRBD > Primary, and you can have it fail over to a DRBD Secondary, but dual primary > is not going to

Re: [DRBD-user] Uncatchable DRBD out-of-sync issue

You forget to CC the list... On 03/25/2013 10:22 AM, Stanislav German-Evtushenko wrote: >> there is a certain risk of data not reaching the hard disk soundly. > It seems so but what is a way to catch the reason? Really not much you can do. Not sure if ZFS is viable and helpful. >> What's your ha

Re: [DRBD-user] Uncatchable DRBD out-of-sync issue

Hi, there is a certain risk of data not reaching the hard disk soundly. This is electronics, bits can get flipped for the dumbest of reasons (yes, space radiation is now more dangerous to your computer than 20 years ago, for one. Or so I've heard.) What's your hard disk stack? ZFS is said to pro

Re: [DRBD-user] Install packages after DRBD startup

Hi, please share your drbd configuration so that we can get a general idea of what your setup looks like. Thanks, Felix On 03/21/2013 08:06 PM, Jeronimo L. Cabral wrote: > Dear, I have a DRBD implementation with two Elastix / CentOS servers. It > works OK. > > But after that I need to install s

Re: [DRBD-user] Migrating data to 4-node setup from node in secondary status

Hi, On 03/17/2013 04:34 AM, Aaron Johnson wrote: > > Due to a unique set of requirements we need to add 2 nodes that are > relatively onsite. > I'm looking for recommendations to migrate from this 2-node setup to a > 4-node configuration without having to pull all 1.5TB over the WAN link I don't

Re: [DRBD-user] 3 node stacked setup and dual primary issues

Well, yes. This (somewhat hacky) construct will indeed allow you to do dual-primary and disaster recovery. However, it is *not* possible to have changes on the DR side make their way back into the dual-master replication. This direction would only be valid if you had the classic setup with the st

Re: [DRBD-user] State change failed: Device is held open by someone

Hi Satyajit, this is an instace of thread confusion. The haresources answer was not actually directed to you, but instead belongs to an old thread that happens to have the same subject. The proposed solutions applies to setups with the obsolete heartbeat cluster stack, which you are (hopefully) n

Re: [DRBD-user] Several cases of hangups found. I have some stack traces to send.

On 03/05/2013 03:11 PM, Lars Ellenberg wrote: >> Well, uh...speaking for myself (but believing that everybody's thinking >> > it), I had kind of expected a bit more. > Sorry to disappoint you :-/ > > There is not much in those stack traces but the fact that > at that point in time apparently drbd_

Re: [DRBD-user] Several cases of hangups found. I have some stack traces to send.

On 02/27/2013 01:45 PM, Lars Ellenberg wrote: > Yep. > Nothing obvious, sorry. > > Lars Well, uh...speaking for myself (but believing that everybody's thinking it), I had kind of expected a bit more. Felix ___ drbd-user mailing list drbd-user@lis

Re: [DRBD-user] drbd pacemaker scst/srp 2 node active/passive question

On 03/01/2013 05:52 PM, Sebastian Riemer wrote: > WTF? Why are you writing about iSCSI? ...said the dude who has yet to keep even one thread on topic. Sorry, couldn't resist ;-) ___ drbd-user mailing list drbd-user@lists.linbit.com http://lists.linbit.c

Re: [DRBD-user] external metadata on ssd vs bcache

On 02/28/2013 12:22 PM, Arnold Krille wrote: >> It's interesting because, normally, writes do not directly translate >> > to head seeks (thanks to dirty pages, caches, NCQ, firmware-level >> > optimization...), and ideally barriers should be disabled (and caches >> > reliable). > If you want secu

Re: [DRBD-user] Device is held open by someone

Uhm, no. Please do not kill the drbd worker threads. They are not holding the device open. There is nothing in the lsof output that suggests that. In fact, the grep only matched the process name here, not the file being held open. Has "fuser -m /dev/drbd7" yielded anything? On 02/27/2013 06:39

Re: [DRBD-user] fence-peer helper broken, returned 20

Good. Have you looked at the script to find out what code 20 actually signifies? On 02/22/2013 02:34 PM, Anoop wrote: > Hi, > Its all configured. > > Thanks > Anoop > > On 2/22/2013 6:56 PM, Felix Frank wrote: >> Hi, >> >> I haven't dabbled in this

Re: [DRBD-user] fence-peer helper broken, returned 20

Hi, I haven't dabbled in this functionality yet, but glancing over http://www.drbd.org/users-guide-8.3/s-heartbeat-dopd.html, it appears that you need to have dopd running, configured etc. Do you? HTH, Felix On 02/22/2013 02:17 PM, Anoop wrote: > Hi, > Thanks for the reply. I have already trie

Re: [DRBD-user] Several cases of hangups found. I have some stack traces to send.

I glanced at it. It looks promising. I'd guess that someone at linbit might be able to spot a deadlock therein. Cheers, Felix On 02/21/2013 07:36 PM, Jesus Climent wrote: > A new D-state with 8.3.15 has appeared in one of the hosts: > > http://dl.dropbox.com/u/20106122/messages > > Anyone care

Re: [DRBD-user] Debian Squeeze drbd with an usb drive, the file goes to read-only after change.

Hi, did your filesystem get mounted read-only? Prune your kernel logs, perhaps something has gone wrong. For example, there are certain LSI RAID/Firmware/Drive/OS combinations that have been known to cause filesystems to get remounted when the RAID became degraded. Regards, Felix On 02/16/2013

Re: [DRBD-user] sync between different drbd versions

Hi, On 02/13/2013 02:20 PM, Christian Gebler wrote: > Hi, > > is it possible or dangerous to sync between two DRBD versions? > > On an older server i have installed: > drbd8-utils 2:8.3.7-1ubuntu2.2 > > The new one: > drbd8-utils 2:8.3.11-0ubuntu1 th

Re: [DRBD-user] Several cases of hangups found. I have some stack traces to send.

Hi, On 02/08/2013 04:32 PM, Jesus Climent wrote: > These are the stack traces: > > http://dl.dropbox.com/u/20106122/trace.D-state.hostname14 > http://dl.dropbox.com/u/20106122/trace.D-state.hostname15 > http://dl.dropbox.com/u/20106122/trace.pri Interesting. I only glanced at those three, but th

Re: [DRBD-user] ProtocollError using on-congestion pull-ahead

Hi, On 02/07/2013 08:58 PM, Holger Steinhaus wrote: > resync-rate 200M; Uhm, what? That's more than your link can do. You want to max this at 30% of your maximum throughput, tops. Otherwise the resync may interfere with on-line write performance. > al-extents 389; Why? This shou

Re: [DRBD-user] Several cases of hangups found. I have some stack traces to send.

Hi, On 02/07/2013 05:45 PM, Jesus Climent wrote: > Hi > > While doing some heavy testing of DRBD 8.3.15 here at Google I have > managed to consistently hung our ganeti clusters as drbd gets in a > very weird inconsistent state, and in some other cases I have had > drbdsetup dead in D-state. > >

Re: [DRBD-user] IRC channel?

Hi, I can reproduce and suspect that it's a problem with this mibbit thing. I heartily recommend installing a slim native client (such as xchat on *nix). HTH, Felix On 02/07/2013 09:41 PM, Dave Kitabjian wrote: > Thanks for the reply! > > > > I'm using a web-based client server one called M

Re: [DRBD-user] File system is running out of space - urgent

Hi, the urgency is appreciated, however - this has *nothing* to do with drbd. The space drbd is leaving usable is 61925464 1k blocks, of which you have used 54821544. drbd internals have nothing to do with this, and you would see the same shenanigans if this was a plain old disk. The most likely

Re: [DRBD-user] Error in formatting DRBD Disk

Hi, On 02/07/2013 05:04 AM, Salman Riaz wrote: > Hello, > > I have successfully deployed the tertiary stacked node. > > [root@n1 ~]# cat /proc/drbd > version: 8.3.13 (api:88/proto:86-96) > GIT-hash: 83ca112086600faacab2f157bc5a9324f7bd7f77 build by > mockbu...@builder10.centos.org

Re: [DRBD-user] State change failed: Device is held open by someone

> View this message in context: > http://drbd.10923.n7.nabble.com/State-change-failed-Device-is-held-open-by-someone-tp11250p17357.html Great. This ugly thing. I've found that it encourages list members to refrain from quoting earlier messages, which I personally find distasteful because I must c

Re: [DRBD-user] Unable to create Metadata for Tertiary node

Hi, On 02/01/2013 06:17 PM, Salman Riaz wrote: > Hello, > > I am unable to create metadata on tertiary node > > # drbdadm --stacked create-md r0-1 > > al_offset 4397912182784 > bm_offset 439969152 > > Found ext3 filesystem > 4294836152 kB data area apparently used > 4294705048 kB left

Re: [DRBD-user] DRBD & Infiniband

Hi, you may want to look at SDP: http://www.drbd.org/users-guide/s-replication-transports.html Sebastian: The "drbd isn't really that great" notion, albeit interesting, strikes me as a little off topic. Cheers, Felix On 01/30/2013 10:24 AM, Sebastian Riemer wrote: > On 29.01.2013 22:32, Steve w

Re: [DRBD-user] DRBD is passing I/O-error to upper layer, but should not

OT: Well nyes - have you tried and issued a mount -o remount,rw for the afflicted virtual partitions? I wouldn't necessarily expect that to work, say, for a disturbed root partition in your scenario, but I *have* seen LSI controllers causing some Linuxes to remount read-only on RAID degradation, a

Re: [DRBD-user] DRBD is passing I/O-error to upper layer, but should not

Hi, On 01/26/2013 06:19 PM, Matthias Hensler wrote: > Jan 26 15:32:21 lisa kernel: block drbd12: IO ERROR: neither local nor remote > data, sector 0+0 > Jan 26 15:32:21 lisa kernel: block drbd9: IO ERROR: neither local nor remote > data, sector 0+0 I'm not at all sure about this, but this does

Re: [DRBD-user] Make resource primary after reboot

Hi, could it be that your boot order puts drbd before a proper networking setup? I.e., drbd starts, but cannot yet connect to its peer, thus just works disconnected? It shouldn't ordinarily do that, but from what little info there is, something in your setup is strange, anyhow. Please note that

Re: [DRBD-user] drbd based logical volume not getting marked active on boot

Hi, On 01/13/2013 10:14 PM, Andy Dills wrote: > I'm trying to setup a gfs2 filesystem on a drbd 'c' volume (to facilitate > migration of openvz containers). > > Everything is working great with the exception of reboots. On reboot, the > filesystem wasn't getting marked as active, and the only

Re: [DRBD-user] Two disk secondary and primary marked DRBD system

>From what I remember from the old thread, your system is too clever for its own good and makes the backing device unusable for you. Shame. If you've got a complete disaster at your hands, you should take a dd clone of your block device anyway. Make sure to strip the drbd metadata (at the end) off

Re: [DRBD-user] Please help... After reboot I'm always getting unresolved split brain (DRBD+OCFS2)

Hi, On 01/24/2013 06:04 PM, Jacek Osiecki wrote: umount: /dev/drbd0: not mounted all: Failure: (127) Device minor not allocated ERROR: Module drbd is in use . That looks very weird. I don't quite remember how to cleanly unmount an ocfs2. Make sure you can start and stop both the FS and DRBD

Re: [DRBD-user] Two disk secondary and primary marked DRBD system

Hi, a similar issue was discussed back in http://www.gossamer-threads.com/lists/drbd/users/20319 In a nutshell, try mount -t ext4. HTH, Felix On 01/19/2013 08:56 AM, Abdelkarim Mateos Sanchez wrote: > Hi. > > Just we like use secondary disc on drbd conf over LVM (LVM logical > volumens on two

Re: [DRBD-user] Please help... After reboot I'm always getting unresolved split brain (DRBD+OCFS2)

On 01/22/2013 05:04 PM, Jacek Osiecki wrote: > [41706.085879] block drbd0: PingAck did not arrive in time. > [41706.085888] block drbd0: peer( Primary -> Unknown ) conn( Connected > -> NetworkFailure ) pdsk( UpToDate -> DUnknown ) > [41706.086007] block drbd0: new current UUID > 62770026DDB5FC9D:1A

Re: [DRBD-user] DRDB stalled and impossible restart, down...

Hi, On 01/11/2013 10:36 AM, Abdelkarim Mateos Sanchez wrote: > Dec 31 17:52:31 pro01 kernel: block drbd1: [drbd1_worker/20189] sock_sendmsg > time expired, ko = 4294961767 > Dec 31 17:52:37 pro01 kernel: block drbd1: [drbd1_worker/20189] sock_sendmsg > time expired, ko = 4294961766 > Dec 31 17:5

Re: [DRBD-user] Tunning DRBD for small writes

Hi, On 01/10/2013 10:09 PM, Lin Zhao wrote: > BTW we are using drbd 8.2 Why? > On Thu, Jan 10, 2013 at 1:08 PM, Lin Zhao > wrote: > > All, > > I'm setting up a drbd setup for my system, but the poor disk write > performance is really throttling my system,

Re: [DRBD-user] Best practice advice needed

Hi, On 12/28/2012 10:49 PM, Andreas Heinlein wrote: > * Split it up into multiple DRBD devices, one for each service, and put > the LVM underneath the DRBD layer to be able to resize when needed. > Would mean several devices would have to be encrypted individually, > which adds complexity, but it'

[DRBD-user] Fwd: Re: Secondary Performance

Personally, I would not use SSDs for that type of deploy. It is best placed to speed up certain processes (swap, scratch pad database location used for indexes or anythi - Original Message ----- From: Felix Frank [mailto:f...@mpexnet.de] Sent: Thursday, December 20, 2012 03:56 AM To: Adam Gor

Re: [DRBD-user] Secondary Performance

On 12/20/2012 08:31 AM, Adam Goryachev wrote: > The primary has 5 x 480G Intel SSD drives in a RAID5 configuration > The secondary has 4 x 2TB WD RE4 drives in a RAID10 configuration Huh. SSD RAID vs. rotaries? That's about as far apart as you can set primary and secondary performance-wise. Have y

Re: [DRBD-user] Moving dm_crypt+LVM to dm_crypt+DRBD+LVM

On 12/12/2012 01:12 PM, Andreas Heinlein wrote: >> >> Seeing as performance is obviously not an issue at all in your setup, I >> disbelieve that you will have to be especially careful about your DRBD >> setup. > Well, depends on what you call 'performance'. How'd you get this idea? > I'd like to be

Re: [DRBD-user] Moving dm_crypt+LVM to dm_crypt+DRBD+LVM

Hi, On 12/12/2012 12:30 PM, Andreas Heinlein wrote: > Hello, > > I am currently planning a migration of a one-machine setups to a > two-machine-cluster. Part of it will be migrating existing data to DRBD, > and I hope you can help me with this. > > The current storage layout looks like this: >

Re: [DRBD-user] What can cause ": I shall become SyncTarget, but I am primary!'

Hi, On 11/29/2012 08:45 PM, John Anthony wrote: > Felix, > > Primary was secondary - Yes! - And Secondary was primary. > > Then we switched roles by: > 1. disconnecting the devices > 2. Explicitly setting old secondary side to primary and old primary side > to secondary. We also swiched the > 3.

Re: [DRBD-user] What can cause ": I shall become SyncTarget, but I am primary!'

Thanks! On 11/29/2012 07:56 PM, John Anthony wrote: Also - Is it safe to invalidate the secondary ? that should initiate a sync, correct ? Well, yes, but... Nov 29 05:39:26 PRDIQ kernel: block drbd42: self 862535653F3D9257:0001:0001:872E75BBB0EABE21 bits:7042712 flags

Re: [DRBD-user] What can cause ": I shall become SyncTarget, but I am primary!'

Hi, On 11/29/2012 05:44 PM, John Anthony wrote: One side is Primary and the other side is Secondary when this message occurs. So not a split brain situation. It's funny you should say that because... But the secondary side has gone through a cycle of Secondary->Primary (validate date)->Secon

Re: [DRBD-user] dedicated network connection

Hi, On 11/25/2012 08:46 PM, marcus wrote: > > Does this mean a second NIC (in each host) on the same subnet, > a second NIC (in each host) on a different, separate switch and subnet, > or a second NIC (in each host) and a configured cross-over cable?? I suggest you trawl the mailing list archive

Re: [DRBD-user] Can't create internal meta-disk

Hi, On 11/26/2012 12:05 AM, marcus wrote: > > Found ext3 filesystem > 5120 kB data area apparently used > 51198400 kB left usable by current configuration QUESTION: huh. So your filesystem is but half the size of your device? Or am I misreading this? Something seems fishy here. > how can I

Re: [DRBD-user] drbd0: State change failed: Device is held open by someone

Hi Dan, please don't forget to CC the list in your replies. On 11/19/2012 08:10 PM, Phillips, Dan wrote: > [root@cali-a ~]# crm configure show > -bash: crm: command not found Ugh, so - heartbeat. :( I don't know heartbeat. But anyhow... > And > > /etc/ha.d/ha.cf: > > Lots of initialization in

Re: [DRBD-user] drbd0: State change failed: Device is held open by someone

On 11/16/2012 10:00 PM, Phillips, Dan wrote: > What is this DRBD used for? Is there a filesystem in it? Or is it used by an > iSCSI target? Something else entirely? > > - I am new to DRBD so I may not fully be able to answer your question. I know > we use DRBD, LVM, snapshot, HA, heartbeat. Ther

Re: [DRBD-user] drbd0: State change failed: Device is held open by someone

Hi, On 11/15/2012 09:02 PM, Phillips, Dan wrote: > Problem: > > The problem is that when performing an HA failover from server A to > server B, a DRBD resource is sometimes not shut down properly on server > A. Several attempts are made to stop the DRBD resource, but finally it > gives up and th

Re: [DRBD-user] Migrate LVM to DRBD+LVM ?

Hi, On 11/14/2012 09:03 PM, Jake Smith wrote: 1. Grow the logical volumes the appropriate amount to have room for the metadata at the end of the device. Estimate the space need (pad it a little if you like) and expand the LV. This method would be my preference. 2. Add another logical volume

Re: [DRBD-user] dual primary + pacemaker issue

On 11/12/2012 11:32 AM, Daniel Grunblatt wrote: >>Why does the drbd start fail? > I wish I knew the answer ;) Then let me rephrase: Are there any drbd and/or pacemaker related logs from right before the shutdown that indicate what's going wrong? Cheers, Felix _

Re: [DRBD-user] dual primary + pacemaker issue

On 11/09/2012 03:30 PM, Daniel Grunblatt wrote: > Problem is that when we want to start the master drbd0 resource in > pacemaker, it fails and then xs02 powers off. Soon after that, the > resource promotes to Master in xs01. I go to the XS02 server, boot up > and the resource is also promoted to Ma

Re: [DRBD-user] Is there any problem using DRBD over LVM?

Hi, On 11/08/2012 12:18 PM, "[SOLTECSIS] Carles Xavier Munyoz Baldó" wrote: > Hi, > We are creating a PROXMOX 2 cluster using DRBD and we have doubts about > which of the next options is better: > (A) DRBD over LVM. > (B) DRBD as a PV for LVM. > > Which is the best option? > > I remember that I

Re: [DRBD-user] Performance puzzle

Hi, On 11/06/2012 05:18 PM, Tor wrote: > the drbd-device is > on a HW-raid5 with eight drives (total of 6.4 TiB) I'm not sure this is a very sound idea, performance-wise. The RAID5 in and of itself should be fine, but you should really consider external metadata IMHO. > It should be close to the

Re: [DRBD-user] GFS2 freezes

On 10/31/2012 10:53 AM, Zohair Raza wrote: > Hi. > > Right now just simulating on vmware, have to setup on real hardware later. Well yes, VMware based fencing won't work in production then. I'd advise to try with IPMI-enabled hardware and go from there, but do mind the caveats as laid out by La

Re: [DRBD-user] GFS2 freezes

Hi, so you're building an active-active cluster via WAN? That requires guts. Do the services run on physical hardware or in virtual machines? In the latter case, it should be quite possible to implement fencing through the virtualization layer (vmware was mentioned earlier in this thread, but not

Re: [DRBD-user] GFS2 freezes

On 10/31/2012 12:02 AM, Lars Ellenberg wrote: >>> Manual fencing is not in any way supported. You must be able to call >>> > > 'fence_node ' and have the remote node reset. If this doesn't >>> > > happen, your fencing is not sufficient. >> > fence_node doesn't work for me >> > >> > fence_node nod

Re: [DRBD-user] GFS2 freezes

Hi, On 10/29/2012 04:19 PM, Zohair Raza wrote: > It worked for me with fencing disabled, Now I still have access to my > gfs2 partition when one node is unplugged oh, it works, but as soon as you have a (short) connectivity issue between your DRBD nodes, your cluster ends up in a split brain situ

Re: [DRBD-user] GFS2 freezes

Hi, On 10/29/2012 02:43 PM, Maurits van de Lande wrote: > I think that adding the following option to the dlm section in cluster.conf > > enable_fencing="0" > > might solve this problem. (but I have not tested this) This will disable > fencing. ...but from what I remember from training and othe

Re: [DRBD-user] Setting up per-VM LVM volumes as individual DRBD mirrors - compatible?

Hi, On 10/28/2012 06:25 PM, Whit Blauvelt wrote: > On Thu, Oct 25, 2012 at 12:34:40PM -0400, Whit Blauvelt wrote: > > Just a note. The edit of the source dev in the kvm xml file did work for one > test VM (once I had the drbd set to primary - duh). But it DID NOT work for > the next one I tried i

Re: [DRBD-user] LCMC display (and other tools) says "up to date" but DRBD is not

On 10/25/2012 05:38 PM, Whit Blauvelt wrote: > My shortcoming was not categorizing what libvirt-KVM-QEMU does as "mounting" > when it's using a direct write to a raw partition. If my current, still > sketchy understanding is right, did I only miss editing the XML file to > change what's mounted by

Re: [DRBD-user] LCMC display (and other tools) says "up to date" but DRBD is not

Mostly what Adam said, but specifically: On 10/25/2012 03:38 PM, Whit Blauvelt wrote: > In the case where that data was a normal > filesystem, that's what DRBD does, right? That's the point of the offer to > duplicate the existing data from primary to secondary? So with a normal > filesystem, you

Re: [DRBD-user] IO Error Logging

Hi, On 10/07/2012 05:46 PM, Andrew Eross wrote: I can change from open vswitch to normal linux bridging with "xe-switch-network-backend bridge" on XenServer, so what the heck let's see if that does some magic here. meh, that seems a little exaggerated to me. Doesn't Xenserver allow you to un-

Re: [DRBD-user] IO Error Logging

On 10/07/2012 05:13 PM, Andrew Eross wrote Very strange! Any ideas, guys? Yeah. I haven't looked at your config, but if DRBD relies on an OVS port for replication, that may be a Bad Idea. HTH, Felix ___ drbd-user mailing list drbd-user@lists.linbi

Re: [DRBD-user] IO Error Logging

On 10/07/2012 03:22 PM, Dan Barker wrote: My system logs them with timestamps. They just happen to be relative to boot time. I dmtime say "dmtime 12345" and see the real time stamp. Yes, most kernels do that. In this particular use case, those relative times would even suffice, because one co

Re: [DRBD-user] IO Error Logging

Hi, On 10/06/2012 03:38 AM, Andrew Eross wrote: Below is what I'm seeing in dmesg. No timestamps? Bummer. Does your system log those via syslog too (in Debian, typicalle /var/log/kern.log)? That log typically has far superior timestamps even. Cheers, Felix _

Re: [DRBD-user] borked split-brain recovery

On 10/03/2012 10:06 PM, mdavid...@allureglobal.co wrote: > in the middle of trying to manually recover from a split-brain, it seems i've > created a little bit of a mess. Well, what steps did you perform on which node to resolve your split-brain? Have you properly told openfiler1 to discard its da

Re: [DRBD-user] Status Mismatch

On 09/19/2012 06:21 PM, J.R. Lillard wrote: > DRBD 8.3.11 with Proxy 3.0.6 Right, sorry for never following up. You have proxy - have you not purchased support for that? (Pardon my ignorance, I was under the impression that the proxy comes with a Linbit contract only.) Cheers, Felix

Re: [DRBD-user] incompatible after-sb-0pri settings

On 09/21/2012 04:52 PM, Lonni J Friedman wrote: > I just discovered that if I run 'drbdadm adjust r0' on the primary a > second time, after running it on the secondary, that fixes everything. > Did I misunderstand, and perhaps the command needs to be run first on > the secondary, and then on the p

Re: [DRBD-user] incompatible after-sb-0pri settings

On 09/21/2012 04:43 PM, Lonni J Friedman wrote: >>> after-sb-1pri discard-secondary; >> > >> > This is you signing up for grotesque data loss under unfortunate >> > circumstances. >> > >> > I advise to never discard anything but zero-changes automatically. > The documentation only l

Re: [DRBD-user] what's the correct procedure for promoting secondary to primary (when primary is down)?

Hi, the scenario is really very common and requires no action aside from those documented everywhere. On 09/21/2012 04:17 PM, Lonni J Friedman wrote: > I'm trying to do a proof of concept, in which I was attempting to > synthetically simulate the failure of the primary. I did this by > bringing

Re: [DRBD-user] incompatible after-sb-0pri settings

On 09/21/2012 01:30 PM, Felix Frank wrote: >> > I don't understand how this is incompatible. At least the >> > documentation doesn't make any mention of what would make this >> > incompatible. > We have yet to see the configuration of the peer node. &

Re: [DRBD-user] incompatible after-sb-0pri settings

Hi, two remarks inline. On 09/20/2012 09:32 PM, Lonni J Friedman wrote: > after-sb-1pri discard-secondary; This is you signing up for grotesque data loss under unfortunate circumstances. I advise to never discard anything but zero-changes automatically. > I don't understand how

Re: [DRBD-user] what's the correct procedure for promoting secondary to primary (when primary is down)?

Hi, On 09/21/2012 01:01 AM, Lonni J Friedman wrote: > I'm running version 8.3.11 on a two node Fedora16-x86_64 setup. I've > read all of the official documentation, but I'm unclear what the > correct procedure is for (manually) promoting the secondary to become > primary when the original primary

Re: [DRBD-user] Status Mismatch

On 09/19/2012 05:10 PM, Dan Barker wrote: > The sync hasn’t finished. It’s at 100%, but still doing cleanup at > end-of-task. When it completes, you’ll see the correct status. > Inconsistent is the VALID status until the sync finishes. When the > progress bar goes away, it’s really done. Check the

Re: [DRBD-user] What to do about read errors on the primary?

On 09/19/2012 08:53 AM, Arnold Krille wrote: > On Monday 17 September 2012 16:04:09 Alan Robertson wrote: >> > I have read errors on the primary side, which caused the secondary to go >> > into an "inconsistent" state. This means that the disk which >> > desperately needs backing up, is no longer

Re: [DRBD-user] stacking drbd 3 high

On 08/28/2012 09:50 PM, Two Spirit wrote: > I'm sorry, I have no clue what the diagram represents nor what the () > signifies in the diagram. I'm still a newbie, could I get layman english > translation please. Hmm, turns out thunderbird displays this quite awkwardly, for one. > > A--B >

Re: [DRBD-user] oracle on drbd failed

On 08/30/2012 09:38 AM, Felix Frank wrote: >> I think you just misunderstood me. The key action for this test is >> > >> > drbdadm disconnect >> > drbdadm primary >> > >> > which simulate the situation that the primary is crashed to test if

Re: [DRBD-user] oracle on drbd failed

On 08/29/2012 05:12 PM, Mia Lueng wrote: > I think you just misunderstood me. The key action for this test is > > drbdadm disconnect > drbdadm primary > > which simulate the situation that the primary is crashed to test if > the oracle can be fail over on secondary node > > drbdadm --discar

  1   2   3   4   5   >