After reboot after update, alot more memory is used

Bug #598849 reported by Ian Corne on 2010-06-26
16
This bug affects 2 people
Affects Status Importance Assigned to Milestone
dpkg (Ubuntu)
Undecided
Unassigned

Bug Description

Binary package hint: dpkg

I'm not sure dpkg is the one to blame.

When booting after an update, 747mb/1000mb ram is used. When i boot normally, only 150mb is used.

It's only the first boot after an update.

This cripples the systems with only 1gb pretty hard.

ProblemType: Bug
DistroRelease: Ubuntu 10.10
Package: dpkg 1.15.7.2ubuntu1
ProcVersionSignature: Ubuntu 2.6.35-6.7-generic 2.6.35-rc3
Uname: Linux 2.6.35-6-generic i686
Architecture: i386
Date: Sat Jun 26 18:33:07 2010
InstallationMedia: Ubuntu 10.10 "Maverick Meerkat" - Alpha i386 (20100601)
ProcEnviron:
 PATH=(custom, no user)
 LANG=en_US.utf8
 SHELL=/bin/bash
SourcePackage: dpkg

Ian Corne (icorne) wrote :
Dave Gilbert (ubuntu-treblig) wrote :

Can you define 'used' - is this in the output of free? top? If it's in this state and you do a top and type a capital M to sort by
memory what's the biggest user?

Dave

Changed in dpkg (Ubuntu):
status: New → Incomplete
Ian Corne (icorne) wrote :

Used as in not available.

This is from my other box, as the laptop had rebooted already.

After update after first reboot:

$ free -m
             total used free shared buffers cached
Mem: 3149 1209 1940 0 27 417
-/+ buffers/cache: 764 2385
Swap: 1906 0 1906

$ cat /proc/slabinfo | pastebinit
http://pastebin.com/ZwwE0J3r

Ian Corne (icorne) wrote :

The memory usage in top/htop ps does not add up to the total amount used shown. I've read that kernel memory/processes is/are not shown in top/htop/ps so that might be where the big user is.

Ian Corne (icorne) wrote :

$ ps -eaf | pastebinit
http://pastebin.com/LGJ9NDRE

Ian Corne (icorne) wrote :

Now with capital F
$ ps -eaF | pastebinit
http://pastebin.com/yvfR9wU2

Ian Corne (icorne) wrote :

To show the difference:

After anoter reboot:

$ ps -eaF | pastebinit
http://pastebin.com/iQJ9HNQD

$ cat /proc/slabinfo | pastebinit
http://pastebin.com/HxPGmSe8

$ free -m
             total used free shared buffers cached
Mem: 3149 821 2328 0 154 431
-/+ buffers/cache: 235 2913
Swap: 1906

Ian Corne (icorne) wrote :
Ian Corne (icorne) wrote :
Ian Corne (icorne) wrote :
Ian Corne (icorne) wrote :
Changed in dpkg (Ubuntu):
status: Incomplete → New
Philip Muškovac (yofel) wrote :

Confirmed here on maverick, first reboot: (that's without even X running)
             total used free shared buffers cached
Mem: 3784952 1564912 2220040 0 34808 195148
-/+ buffers/cache: 1334956 2449996

second reboot:
             total used free shared buffers cached
Mem: 3784952 574988 3209964 0 59636 195804
-/+ buffers/cache: 319548 3465404

looks sane again. (Using amd64)

Changed in dpkg (Ubuntu):
status: New → Confirmed

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

The bootcharts from my boots.
The only apparent difference to me are the ureadahead and fsck run
times. (Oh, and failsaveX, but I doubt that has anything to do with this)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkwn+uAACgkQIVvPUU2bvLTimACcDQNBHiqx60fcm/2cNOvYMcte
ckAAn04G8nsxeT7OWo7i/Rts4j+1XEdJ
=5Eoo
-----END PGP SIGNATURE-----

Ian Corne (icorne) wrote :

Another day, another kernel update, no changes. Still eating memory after first reboot

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers