Server is crashing with a 'merced' call trace from vmcore.
Issue
- Server crashed repeatedly, with no obvious errors in the system logs.
- Analyzing a vmcore the following call trace was seen:
Call Trace:
[<ffffffff8827350e>] :mercd:qMsgPoolFree+0x10/0xf9
[<ffffffff88273657>] :mercd:native_freemsg+0x14/0x31
[<ffffffff88285fbd>] :mercd:mercd_copyfromBuffers+0xe1/0x214
[<ffffffff88270ea8>] :mercd:mercd_read+0x0/0x5d
[<ffffffff88286cfc>] :mercd:mercd_readQueue+0x101/0x149
[<ffffffff8008be71>] default_wake_function+0x0/0xe
[<ffffffff88270ef8>] :mercd:mercd_read+0x50/0x5d
[<ffffffff800f7ae5>] compat_do_readv_writev+0x1ba/0x2b4
[<ffffffff800f6370>] sys_epoll_wait+0x3b8/0x3f9
[<ffffffff8002e622>] sys_fcntl+0x2d0/0x2dc
[<ffffffff800f7ff2>] compat_sys_fcntl64+0x2c9/0x381
[<ffffffff800f7cb2>] compat_sys_readv+0x58/0x77
[<ffffffff8006149d>] sysenter_do_call+0x1e/0x6a
Environment
- Red Hat Enterprise Linux 5.4
mercedmodule forcefully loaded.
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.