Hello,
I have just setup 4 redhat 8.0 servers, all of which have 2 NIC's. On 2 of
the servers both NIC's are up and running, the other 2 servers are only,
currently, using 1 NIC.
On the servers with 2 NIC's running the top command is giving some very odd
figures for the memory usage of some processes:
12:23pm up 6 days, 19:46, 1 user, load average: 0.02, 0.02, 0.00
62 processes: 60 sleeping, 2 running, 0 zombie, 0 stopped
CPU states: 0.0% user, 0.0% system, 0.0% nice, 100.0% idle
Mem: 1031620K av, 391052K used, 640568K free, 0K shrd, 77660K buff
Swap: 2097112K av, 924K used, 2096188K free 195840K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
1 root 15 0 888 884 512 S 0.0 0.0 0:04 init
2 root 15 0 0 0 0 SW 0.0 0.0 0:00 keventd
3 root 34 19 0 0 0 SWN 0.0 0.0 0:00 ksoftirqd_CPU0
6 root 25 0 0 0 0 SW 0.0 0.0 0:00 bdflush
4 root 15 0 0 0 0 SW 0.0 0.0 0:00 kswapd
5 root 15 0 0 0 0 SW 0.0 0.0 0:00 kscand
7 root 15 0 0 0 0 SW 0.0 0.0 0:00 kupdated
8 root 25 0 0 0 0 SW 0.0 0.0 0:00 mdrecoveryd
14 root 15 0 0 0 0 SW 0.0 0.0 0:01 kjournald
103 root 15 0 0 0 0 SW 0.0 0.0 0:00 kjournald
475 root 15 0 28.7G 28G 2560M S 0.0 1256.7 0:03
dead_gateway_de
487 root 15 0 672 672 544 S 0.0 0.0 0:01 syslogd
492 root 15 0 548 548 456 S 0.0 0.0 0:00 klogd
501 rpc 21 0 576 572 460 S 0.0 0.0 0:00 portmap
564 nscd 15 0 968 968 800 S 0.0 0.0 0:03 nscd
588 named 25 0 3908 3908 3072 S 0.0 0.3 0:19 named
620 root 15 0 672M 672M 912 S 0.0 66.7 0:05 xinetd
635 ntp 15 0 2036 2036 1692 S 0.0 0.1 0:00 ntpd
649 lp 15 0 78776 76M 1028 S 0.0 7.6 0:00 lpd
As can be seen the system is pretty much idle - I am the only one using it
at the moment. However, the 'dead_gateway_detection' process is using 28GB
of memory! The system only has 1GB of main memory and 2GB of swap. The
'xinetd' process is using well over 600MB of memory and the printer process
'lpd' is using 76MB! Others, not shown, are sshd at 87MB and exim at 94MB -
we run exim rather than sendmail.
The dead_gateway_detection process is a shell script started at boot time to
monitor the NIC's and adjust the IP routing if one of them fails. It is a
pretty short script, so I see no way that it can use 28GB of memory.
Does anyone have any ideas about this? A similar thing is seen on the other
server with both NIC's in use. However both servers seem to be running fine.
The free command shows:
total used free shared buffers cached
Mem: 1031620 391056 640564 0 77660 195844
-/+ buffers/cache: 117552 914068
Swap: 2097112 924 2096188
so it seems that the memory is okay, but that 'top' is showing wrong figures.
The top command comes from the procps-2.0.7-25 rpm.
Thanks,
John.
------------------------------------------------------------------------
John Horne, University of Plymouth, UK Tel: +44 (0)1752 233914
E-mail: jhorne@plymouth.ac.uk
PGP key available from public key servers
--
Psyche-list mailing list
Psyche-list@redhat.com
https://listman.redhat.com/mailman/listinfo/psyche-list
