Dotaz: Po poslednim vetsim updatu systemu mi neco zere pamet

[Strycek Fido]

Dobry den!

Vim ze se to sice nema, ale 18tyho jsem upgradoval vice balicku na mym debianim serveru najednou (Holt jsem pravovernej Gentooista ). Ale od ty doby mi neco soustavne pozira pamet po 4kB strankach v intervalech tak 3-10s dlouhych. Pamet se cpe do bufferu ne do aplikaci. Kdyz se zaplni cela (u notebooku pentium 166MMX s 64MB RAMeti to jde vcelku snadno), pusti se to do swapu.

Driv nez budu mit cas se tim doopravdy prat, chtel bych se zeptat, jestli se s tim uz nekdo nahodou nesetkal..(?). A i kdyz ne, tak mi muzete alespon pomoct tipovat, kterej z nasledujicich updatu by to moh' byt:

bzip2 1.0.2-8 -> 1.0.2-11
console-common 0.7.47 -> 0.7.55
console-data 2002.12.04dbs-46 -> 2002.12.04dbs-50
console-tools 1:0.2.3dbs-55 -> 1:0.2.3dbs-60
hpoj 0.91-4 -> 0.91-9
initrd-tools 0.1.81.1 -> 0.1.84
iputils-ping 3:20020927-2 -> 3:20020927-3
libbz2-1.0 1.0.2-1 -> 1.0.2-11
libc6 2.3.5-6 -> 2.3.5-8
libc6-dev 2.3.5-6 -> 2.3.5-8
libconsole 1:0.2.3dbs-55 -> 1:0.2.3dbs-60
libgcrypt11 1.2.0-4 -> 1.2.2-1
libglib2.0-0 2.4.8-1 -> 2.8.3-1
libperl5.8 5.8.4-3 -> 5.8.7-9
libsensors3 2.8.8-7 -> 1:2.9.2-6
libsnmp-base 5.1.2-6 -> 5.2.1.2-4
libusb-0.1-4 2:0.1.10a-21 -> 2:0.1.10a-22
mc 1:4.6.0-4.6.1-pre1-3 -> 1:4.6.1-1
net-tools 1.60-10 -> 1.60-17
netkit-inetd 0.10-10 -> 0.10-10.2
perl 5.8.4-3 -> 5.8.7-9
perl-base 5.8.4-3 -> 5.8.7-9
perl-modules 5.8.4-3 -> 5.8.7-9
samba-common 3.0.14a-6 -> 3.0.20b-2+b1
smbfs 3.0.14a-6 -> 3.0.20b-2+b1
tar 1.14-2 -> 1.15.1-2
tcpd 7.6.dbs-6 -> 7.6.dbs-8
traceroute 1.4a12-19 -> 1.4a12-20
coreutils 5.2.1-2 -> 5.2.1-2.1
cpio 2.5-1.1 -> 2.6-9
diff 2.8.1-7 -> 2.8.1-11
e2fslibs 1.37-2 -> 1.38-2
e2fsprogs 1.37-2 -> 1.38-2
file 4.12-1 -> 4.15-2
findutils 4.1.20-5 -> 4.2.26-2
grep 2.5.1.ds1-4 -> 2.5.1.ds2-4
gzip 1.3.5-9 -> 1.3.5-12
libdb1-compat 2.1.3-7 -> 2.1.3-8
libgnutls11 1.0.16-13.1 -> 1.0.16-14
libmagic1 4.12-1 -> 4.15-2
libncurses5 5.4-9 -> 5.5-1
libncurses5-dev 5.4-9 -> 5.5-1
libsasl2 2.1.19-1.5 -> 2.1.19-1.7
libssl0.9.7 0.9.7e-2 -> 0.9.7g-5

Zkousel jsem povypinat vsechny daemony pro tiskarny, scanner, jabber, teamspeak a dalsi, ale zralo to dal. :-/

Ja osobne bych si bez hlubsich uvah tipnul na coreutils, e2fslibs nebo net-tools. A co vy?

Odpovědi

[Michal Marek (twofish)]

Ani jeden z těch tří kandidátů neběží jako daemon, takže těžko bude zabírat paměť. Spíš pošli výpisy free.

[georgewh]

Skus top a potom [Shift]+[m]...

[Strycek Fido]

Zkusil, zkusil. Tady vidite zkraceny vysledek:

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 1268 ftpuser    9   0  2456 2452 2272 S  0.0  3.9   0:00.22 sshd
 1265 root       9   0  2400 2396 2220 S  0.0  3.8   0:00.39 sshd
  698 jabber     9   0  2176 2172 1468 S  0.0  3.5   0:21.39 jabberd
  493 root       9   0  1956 1924 1212 S  0.0  3.1   0:00.37 cupsd
 1269 ftpuser    9   0  1728 1728 1156 S  0.0  2.8   0:00.06 bash
 1260 root       9   0  1528 1528 1160 S  0.0  2.4   0:00.17 bash
 1272 root       9   0  1512 1512 1164 S  0.0  2.4   0:00.14 bash
  700 jabber     9   0  1652 1420 1308 S  0.0  2.3   0:00.04 jabberd

[Strycek Fido]

Jo, prominte. Uz posilam. free Vam toho asi ale moc nerekne. Tak tedy alespon na porovnani:

             total       used       free     shared    buffers     cached
Mem:         62708      56916       5792          0       9008      34540
Low:         62708      56916       5792
High:            0          0          0
-/+ buffers/cache:      13368      49340
Swap:        92696       1492      91204
Total:      155404      58408      96996

Normalne byva pouzita pamet tusim tak neco kolem 33MB (mozna vic) a swap nula. Buffers netusim. Ted uz se pomalinky zase zacina plnit swap. Tedy je jeste nekde nejaka pamet asi ztracena, ale zadne zombie. Kdyz daemoni nic nedelaj', tak vsechno pribyva vazne jenom do tech bufferu.

Procesy:

F S   UID   PID  PPID  C PRI  NI ADDR SZ WCHAN  TTY          TIME CMD
4 S     0     1     0  0  69   0 -   348 select ?        00:00:04 init
1 S     0     2     1  0  68   0 -     0 contex ?        00:00:00 keventd
1 S     0     3     1  0  69   0 -     0 apm_ma ?        00:00:00 kapmd
1 S     0     4     1  0  79  19 -     0 ksofti ?        00:00:00 ksoftirqd_CPU0
1 S     0     5     1  0  69   0 -     0 kswapd ?        00:00:01 kswapd
1 S     0     6     1  0  69   0 -     0 bdflus ?        00:00:00 bdflush
1 S     0     7     1  0  69   0 -     0 kupdat ?        00:00:00 kupdated
1 S     0     8     1  0  69   0 -     0 kjourn ?        00:00:01 kjournald
1 S     0   116     1  0  69   0 -     0 kjourn ?        00:00:03 kjournald
1 S     0   173     1  0  69   0 -     0 ?      ?        00:00:00 khubd
1 S     0   458     1  0  69   0 -   362 select ?        00:00:00 syslogd
5 S     0   461     1  0  69   0 -   487 syslog ?        00:00:00 klogd
1 S   103   478     1  0  69   0 -   661 select ?        00:00:00 ptal-mlcd
1 S   103   481     1  0  69   0 -   829 select ?        00:00:00 ptal-printd
1 S   103   484     1  0  69   0 -   829 wait_f ?        00:00:00 ptal-photod
5 S     0   493     1  0  69   0 -  1245 select ?        00:00:00 cupsd
5 S     0   512     1  0  69   0 -   356 select ?        00:00:00 inetd
1 S     0   559     1  0  69   0 -   353 select ?        00:00:00 cardmgr
5 S     0   618     1  0  69   0 -  1050 select ?        00:00:00 sshd
1 S  1000   654     1  0  69   0 -  3978 nanosl ?        00:00:01 server_linux
1 S  1000   655   654  0  69   0 -  3978 poll   ?        00:00:00 server_linux
1 S  1000   656   655  0  69   0 -  3978 poll   ?        00:00:27 server_linux
1 S  1000   657   655  0  69   0 -  3978 poll   ?        00:00:39 server_linux
1 S  1000   658   655  0  69   0 -  3978 select ?        00:01:26 server_linux
1 S  1000   659   655  0  69   0 -  3978 select ?        00:00:02 server_linux
1 S  1000   660   655  0  69   0 -  3978 nanosl ?        00:00:00 server_linux
1 S  1000   663     1  0  69   0 -  1177 poll   ?        00:00:00 opendchub
1 S  1000   667   663  0  69   0 -  1181 poll   ?        00:00:00 opendchub
0 S     0   674     1  0  69   0 -   346 read_c tty2     00:00:00 getty
0 S     0   675     1  0  69   0 -   346 read_c tty3     00:00:00 getty
0 S     0   676     1  0  69   0 -   346 read_c tty4     00:00:00 getty
0 S     0   677     1  0  69   0 -   346 read_c tty5     00:00:00 getty
0 S     0   678     1  0  69   0 -   346 read_c tty6     00:00:00 getty
4 S   101   698     1  0  69   0 -  1231 select ?        00:00:21 jabberd
1 S   101   700   698  0  69   0 -   959 pipe_w ?        00:00:00 jabberd
4 S     0  1260     1  0  69   0 -   642 wait4  tty1     00:00:00 bash
0 S     0  1264  1260  1  71   0 -   498 select tty1     00:01:00 top
4 S     0  1265   618  0  69   0 -  1767 ?      ?        00:00:00 sshd
5 S  1001  1268  1265  0  71   0 -  1767 select ?        00:00:00 sshd
0 S  1001  1269  1268  0  69   0 -   781 wait4  pts/0    00:00:00 bash
4 S     0  1272  1269  0  69   0 -   637 wait4  pts/0    00:00:00 bash
4 R     0  1274  1272  0  78   0 -   732 -      pts/0    00:00:00 ps

Tu taky nevidim nic zvlastniho oproti normalnu.

Existuje nejaky prikaz pro konkretnejsi vypis, co alokuje ty buffery?

[Ruža Becelin]

http://www.root.cz/clanky/kam-se-podela-pamet/

[?]

raději

[Ruža Becelin]

sorry, kopiroval jsem vickrat po sobe :-]]]]