I have been scouring the net for my specific issue. All the responses
suggest that either the file permissions are wrong or that the ulimit
is set too low. I know that neither of these is my problem because
Apache starts at boot and sometimes restarts while the server is
running. While it is running I seem to be able to kill -1 as often as
I want.
However, if the server stops for any reason, there is a good chance
that it won't restart and gives me the "unable to open logs" message.
I am running NetBSD 5.1. Here is some info from sysctl relating to
number of files:
kern.maxfiles = 52364
proc.curproc.rlimit.filesize.soft = unlimited
proc.curproc.rlimit.filesize.hard = unlimited
An fstat | wc -l gives numbers an order of magnitude below maxfiles.
Currently, even with Apache running, it is sitting at 7638. When I try
to restart my ulimit is unlimited.
Here is my memory usage currently.
MemTotal: 16252760 kB
MemFree: 14894768 kB
MemShared: 0 kB
Buffers: 304496 kB
Cached: 799904 kB
SwapTotal: 8389548 kB
SwapFree: 8389548 kB
Here is more information about my system:
NetBSD 5.1 (VEX) #0: Sun Oct 21 09:19:41 EDT 2012
door@xxxxxxxxxxxxxx:/usr/obj/sys/arch/amd64/compile.amd64/VEX
total memory = 16382 MB
avail memory = 15868 MB
timecounter: Timecounters tick every 10.000 msec
timecounter: Timecounter "i8254" frequency 1193182 Hz quality 100
SMBIOS rev. 2.31 @ 0xdc010 (57 entries)
HP ProLiant DL140 G3 (436602-001)
mainbus0 (root)
cpu0 at mainbus0 apid 0: Intel 686-class, 2992MHz, id 0xf64
cpu1 at mainbus0 apid 4: Intel 686-class, 2992MHz, id 0xf64
cpu2 at mainbus0 apid 2: Intel 686-class, 2992MHz, id 0xf64
cpu3 at mainbus0 apid 6: Intel 686-class, 2992MHz, id 0xf64
cpu4 at mainbus0 apid 1: Intel 686-class, 2992MHz, id 0xf64
cpu5 at mainbus0 apid 5: Intel 686-class, 2992MHz, id 0xf64
cpu6 at mainbus0 apid 3: Intel 686-class, 2992MHz, id 0xf64
cpu7 at mainbus0 apid 7: Intel 686-class, 2992MHz, id 0xf64
As for log permissions, here are all the files and directories,
including parents, with today's timestamps.
-rw-r--r-- 1 root wheel 5955946 Feb 25 00:00 suexec.log.2013.02.25
drwxr-xr-x 2 root wheel 3584 Feb 25 00:00 .
drwxr-xr-x 4 root wheel 7168 Feb 25 00:00 ..
-rw-r--r-- 1 root wheel 23050427 Feb 25 00:00 access.2013.02.25
-rw-r--r-- 1 root wheel 7462145 Feb 25 00:00 error.2013.02.25
-rw-r--r-- 1 root wheel 1825540 Feb 25 11:50 error
-rw-r--r-- 1 root wheel 3246059 Feb 25 11:51 suexec.log
-rw-r--r-- 1 root wheel 5245858 Feb 25 11:51 access
As you can see I am running suExec but this problem happened before I
implemented that. In fact, suExec allowed me to run one server instead
of multiple ones. I was hoping that it would solve my problem by
reducing resource usage but no joy.
I can't imaging what other information I can supply here but if you
need more please let me know. This machine takes five minutes to
reboot so I would really like a way that does not involve booting.
Any help really appreciated.
--
D'Arcy J.M. Cain
System Administrator, Vex.Net
http://www.Vex.Net/ IM:darcy@xxxxxxx
Voip: sip:darcy@xxxxxxx
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx
For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|