[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[cobalt-users] No load, all memory used after improper shutdown



Hi all,

Two of my RaQ4i's were unexpectedly cycled yesterday (power blink, bad UPS).
Neither were properly shut down, both came back up immediately. Now they
both are using swap memory, though neither has any load whatsoever. Here's
the top output from one, a 4i with 512MB:

----
  7:21pm  up 1 day,  5:58,  1 user,  load average: 0.02, 0.02, 0.00
36 processes: 35 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  0.1% user,  1.3% system,  0.0% nice, 98.4% idle
Mem:   517188K av,  509068K used,    8120K free,   98268K shrd,  393572K
buff
Swap:  131532K av,     308K used,  131224K free                   41284K
cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT  LIB %CPU %MEM   TIME COMMAND
19763 httpd      0   0  7904 7904  6752 S       0  0.0  1.5   0:00 httpd
 4724 httpd      0   0  7904 7904  6736 S       0  0.0  1.5   0:00 httpd
 4728 httpd      0   0  7904 7904  6740 S       0  0.0  1.5   0:00 httpd
---

I've sorted the output by memory so no need to show the balance. For
reference this is what I normally get after a clean reboot:

---
 11:55am  up 16 min,  1 user,  load average: 0.00, 0.03, 0.05
35 processes: 34 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  0.5% user,  0.5% system,  0.0% nice, 98.8% idle
Mem:   517188K av,   66244K used,  450944K free,   88864K shrd,    8876K
buff
Swap:  131532K av,       0K used,  131532K free                   29604K
cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT  LIB %CPU %MEM   TIME COMMAND
  565 httpd      0   0  7316 7316  7056 S       0  0.0  1.4   0:00 httpd
  566 httpd      0   0  7308 7308  7056 S       0  0.0  1.4   0:00 httpd
  567 httpd      0   0  7308 7308  7056 S       0  0.0  1.4   0:00 httpd

---

Any idea what the deal is? MRTG shows (obviously) swap being allocated after
the reboot on both machines, but before this event neither had even come
close to showing all the physical RAM used.

The archives suggest that there's no way to remedy the situation without
another reboot. While nothing's "wrong" per se, I'm not comfortable with
things not being "normal".

Thanks for your comments.

BT