Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Sun, Oct 26, 2014 at 03:17:39PM +0000, Oren wrote: > Hi, > We are experiencing allot of filesystem issues, for example: > kernel: EXT3-fs error (device drbd0): ext3_free_blocks_sb: bit already cleared for block 5776787 > These occur randomly after several consequative failovers between the cluster nodes. > We are using drbd with ext3 filesystem with a write-through controller. > Are there any preferences what elevator we should use? > According to http://www.drbd.org/users-guide/s-latency-tuning.html > both cfq/deadline are fine with preference to deadline. > But both re-order writes, right? Yes. > possibly breaking file system/journal on standby. Not at all. Unless you configure DRBD for "no disk drain", so don't do that. We analyse the incoming request stream, automatically detect re-order-domains, and separate those with in-drbd-wire-protocol "drbd barrier" packets. > Please advise, > > Setup > drbd with ext3 filesystem > write-through controller > Two nodes cluster > Physical Appliances > CentOS 6.2 with 2.6.32-220 > kmod-drbd83-8.3.12-1.el6.i686 Please use drbd 8.4. You will be so much more happy. -- : Lars Ellenberg : LINBIT | Your Way to High Availability : DRBD/HA support and consulting http://www.linbit.com DRBD® and LINBIT® are registered trademarks of LINBIT, Austria. __ please don't Cc me, but send to list -- I'm subscribed