Messages in this thread | | | Date | Sat, 15 Jan 2000 16:38:45 -0800 (PST) | From | Robert Dinse <> | Subject | Strange NFS (userland) problem... |
| |
Recently I moved a few virtual domains from one web server to another for load balancing reasons.
Not wanting to make the customer look somewhere else for logs, I NFS mounted the directory where the logs reside to the new machine that the domains were on so that they could continue to log in the same location.
I ran into some very strange behavior; logs would grow from zero bytes up to about 4k, only if you cat the file it's all nulls, and then recycle to zero. It seems that when Apache's many processes write to logs there is no synchronization or something.
Moving the logs to a local partition fixed the problem. I could copy files on the NFS mounted partition and they would copy fine. I ended up relocating all of the logs to a local partition and that worked fine.
The machines are presently running Linux 2.2.14, I tried reverting back to 2.2.13 just on the off-chance that it was a newly introduced bug, but exactly the same behavior was present under 2.2.13.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |