[Lucid] Load average seems wrong on 2.6.32-30-server 64-bit

Bug #755369 reported by Gavin McCullagh
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Won't Fix
Undecided
Unassigned

Bug Description

Hi,

since upgrade to lucid from hardy this week, our load average never appears to go below 1, even though the server is really not very busy.

As an example:

gavinmc@magni:~$ vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r b swpd free buff cache si so bi bo in cs us sy id wa
 0 0 0 753580 191860 2733844 0 0 41 3 10 7 4 0 96 0
 1 0 0 747372 191860 2734288 0 0 624 0 518 337 10 0 89 0
 0 0 0 746752 191860 2734732 0 0 512 0 523 379 6 0 94 0
 1 0 0 746132 191860 2735372 0 0 384 0 489 258 4 0 96 0
 0 0 0 745512 191860 2735884 0 0 640 0 457 277 9 0 91 0
 0 0 0 745016 191860 2736304 0 0 384 0 333 261 5 0 95 0
 2 0 0 744520 191876 2736644 0 0 384 72 450 272 4 0 96 0
 0 0 0 743900 191876 2737036 0 0 512 0 482 308 6 0 94 0
 0 0 0 743652 191876 2737420 0 0 256 0 340 249 3 0 97 0
 1 0 0 743156 191876 2738060 0 0 640 0 450 287 5 0 95 0
 0 0 0 742412 191876 2738736 0 0 512 0 470 256 7 0 93 0
 1 0 0 742412 191876 2738956 0 0 256 0 340 251 2 0 98 0
 0 0 0 741792 191884 2739468 0 0 640 20 455 374 7 0 93 0
 0 0 0 741296 191884 2740144 0 0 512 0 387 268 8 0 92 0
 0 0 0 741048 191884 2740236 0 0 384 0 437 257 4 0 96 0
 0 0 0 740304 191884 2740876 0 0 512 0 408 265 4 0 96 0
 0 0 0 739932 191884 2741132 0 0 384 0 485 266 6 0 94 0
 0 0 0 739436 191892 2741640 0 0 384 20 454 290 6 0 94 0
 1 0 0 739436 191892 2741896 0 0 384 0 548 363 4 0 96 0
^C
gavinmc@magni:~$ uptime
 10:01:03 up 4 days, 10:08, 1 user, load average: 1.38, 1.26, 1.39

Needless to say load average is an average over minutes so this isn't enough information. However, I'll attach a munin graph showing that it seems to be asymptoted above 1. You can also see that prior to the upgrade, the load average regularly fell below zero. This new values seems very unlikely. I have a similar munin graph from another Lucid server (albeit running the -generic- 32-bit kernel not the -server- 64-bit) which regularly goes below zero.

Gavin

Revision history for this message
Gavin McCullagh (gmccullagh) wrote :
Revision history for this message
Gavin McCullagh (gmccullagh) wrote :
Revision history for this message
Gavin McCullagh (gmccullagh) wrote :
Revision history for this message
Gavin McCullagh (gmccullagh) wrote :
Revision history for this message
Gavin McCullagh (gmccullagh) wrote :
Brad Figg (brad-figg)
Changed in linux-meta (Ubuntu):
status: New → Confirmed
Brad Figg (brad-figg)
affects: linux-meta (Ubuntu) → linux (Ubuntu)
Revision history for this message
Po-Hsu Lin (cypressyew) wrote :

Closing this bug with Won't fix as Lucid is no longer supported.
Please feel free to open a new bug report if you're still experiencing this on a newer release (Bionic 18.04.3 / Disco 19.04)
Thanks!

Changed in linux (Ubuntu):
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.