[DRBD-user] DRBD + RHCS - Failover not working

Hi All, I have a working 2 node RHCS + DRBD (ext3 filesystem replication) instance up and running. I can relocate the service just fine but when I fail the service (kill a process or power off the VM), the service fails to relocate to the other node. Please help Does ANYONE have a workin

[DRBD-user] allow doubal primary and the notice

Dear all, sometimes, I found "drbdadm secondary resource" blocks or does not works. For example, when the primary "Diskless", "drbdadm secondary resource" will not works. So, the standby can't become DRBD primary. I spent several hours to solve this problem. At last, I think "a

[DRBD-user] Kernel panic - DRBD 8.3.6 and RHEL 5.4 x64 xen

I just installed DRBD 8.3.6, but when starting the module, it immediately causes a kernel panic. Starting DRBD resources: [ d(storage) Unable to handle kernel paging request at 81018341e03c RIP: [] :drbd:drbd_md_read+0xd7/0x236 PGD 183026063 PUD 0 Oops: [1] SMP last sysfs file: /mod

Re: [DRBD-user] Replication over Fiber Links

Hi Mark. I suspected the same, but our storage guy was hoping that there was some way of having the fiber cards talking to each other without the need for the tcpip encapsulation, therefore reducing the overhead Jay On Dec 8, 2009 4:20pm, Mark Watts wrote: On Tue, 2009-12-08 at 11:38 +00

Re: [DRBD-user] Replication over Fiber Links

On Tue, 2009-12-08 at 11:38 +, Jason Fitzpatrick wrote: > Hi all > > Just wondering if it is possible to replicate DRBD over a fiber link.. > > we have 2 datacenters connected by multiple "dark" fiber links, and > would like to have DRBD replication over said links as apposed to > network lin

Re: [DRBD-user] LVM crash maybe due to a drbd issue

Hi, the drbd16 device is indeed blocking lvchange : z2-3:~# lsof /dev/drbd16 COMMANDPID USER FD TYPE DEVICE SIZE/OFFNODE NAME lvchange 4969 root5u BLK 147,16 0t0 1328107 /dev/drbd16 what is the best way to stop it ? (drbdsetup down won't work because lvchange use it). C

[DRBD-user] Replication over Fiber Links

Hi all Just wondering if it is possible to replicate DRBD over a fiber link.. we have 2 datacenters connected by multiple "dark" fiber links, and would like to have DRBD replication over said links as apposed to network links. Thanks a mill Jay ___ d

Re: [DRBD-user] block drbd0: bio would need to, but cannot, be split error on physical device

> no, it's default kernel from Redhat 5.4 > kernel-2.6.18-164.el5 [vchep...@ashlin02 ~]$ rpm -qa|egrep "kernel|drbd" kernel-devel-2.6.18-164.el5 drbd-8.3.5-3 kernel-headers-2.6.18-164.el5 kernel-2.6.18-164.el5 drbd-km-2.6.18_164.el5-8.3.5-3 ___ drbd-use

Re: [DRBD-user] block drbd0: bio would need to, but cannot, be split error on physical device

--- On Tue, 12/8/09, Lars Ellenberg wrote: > uh. > I'd like to know your exact kernel version. > If not a "standard" kernel, I'd need your kernel config. > no, it's default kernel from Redhat 5.4 kernel-2.6.18-164.el5 ___ drbd-user mailing list drbd-

Re: [DRBD-user] block drbd0: bio would need to, but cannot, be split error on physical device

On Mon, Dec 07, 2009 at 07:26:46PM -0800, Vadym Chepkov wrote: > I never specified any block size or modified any non-default > parameters during compilation. Should I compile drbd differently and > include some flags to reduce per bio size? You did nothing wrong. It is DRBD that does some stupid