Hi!
This is a strange:
2314 ? SW 0:00 (postmaster)
2535 ? S 0:00 (postmaster)
2653 ? SW 0:00 (postmaster)
3028 ? SW 0:00 (postmaster)
3776 ? S 1:40 /usr/comset/dbase/bin/postmaster -i -D/usr/comset/dbase/dat
30334 ? SW 0:00 (postmaster)
31904 ? SW 0:00 (postmaster)
^^ Sleep and NO ONE page resident (all pages of process is in swap).
All (postmaster)s are swapped and never be active. Looks like dieing orphans.
When I kill them as:
ps ax|grep \(postmaster\)|tr -d ' '|cat -d'?' -f1|xargs -n1 kill
All orphans killed. Nothing changed! active postmaster still works..
Once I catch strange situation:
3776 ? S 1:40 \_ /usr/comset/dbase/bin/postmaster -i -D/usr/comset/d
31293 ? S 0:07 \_ /usr/comset/dbase/bin/postgres 194.226.196.17 p
31571 ? S 0:01 \_ /usr/comset/dbase/bin/postgres 194.226.196.17 p
31661 ? D 0:00 \_ /usr/comset/dbase/bin/postmaster -i -D/usr/coms
^ This means uninterruptible sleep
A second later this process was daed successfully but may be such process
falled into swap as I show above.
A few time before this one process serve my connection via psql:
31644 ? S 0:00 \_ /usr/comset/dbase/bin/postgres 194.226.196.20 s
Hmmm... The orphans occupied swap space and then there is many orphans the
postmaster becomes cancell connection :-(
SKiller
--------------------------
Sergei Keler
WebMaster of "ComSet"
E-Mail: [EMAIL PROTECTED]
http://www.comset.net
--------------------------
************
