Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi,
I'm not sure if this is a bug or not. I'm using DRBD (v8.0.6). At most
of the time, DRBD works well. But after about 100 failovers, some files
are corrupted (These files have the contents of other files, it could be
caused by messed i-nodes or duplicate i-nodes). One thing I want to
mention is that during the Primary node is going down, an application is
writing data to an file in the replication partition (/repl). During
this period, this primary node is umounting /repl too.
As a workaround, "drbdadm invalidate-remote all" is added right before
/repl is umounted. The purpose is to force sync before failover. After
this workaround, the file corruption has not been seen even after about
160 failovers. At this time, I don't have data more than about 160
failovers.
As far as I know, DRBD will do real-time sync. Is force sync ("drbdadm
invalidate-remote all") a must-have step in my case? Or is there a bug
in DRBD? Are there any people reported similar issue before?
Your timely reply is highly appreciated.
Thanks,
Zhao Que
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20100527/e9fe2277/attachment.htm>