swapfile thrashing in tiger after updates

[superbovine]

I've seem to noticed after the last update was pushed or the last itunes update (i don't remember which) that my swapfile is getting thrashed majorly hard. I ran "top" and nothing looked out of the ordinary. the trashing got so bad, I was waiting for my hd to finish its vm swap while my window animations froze midway. the only way to solve my problem right now is to delete the /var/vm/swapfile* and reboot file which isn't a very good solution. the weird thing is sometime to get past the slash screen on boot sometimes I have to reset my PRAM.

My other guess is my hard disk maybe is going bad, which means it a bitch to replace the hd.

I have powerbook 12" dvi with 768 mb ram and 13.7/37.7 GB HD

anybody got any ideas?

baseline vm_stats:

Mach Virtual Memory Statistics: (page size of 4096 bytes)
Pages free: 125381.
Pages active: 22274.
Pages inactive: 32151.
Pages wired down: 16802.
"Translation faults": 288009.
Pages copy-on-write: 15193.
Pages zero filled: 135875.
Pages reactivated: 0.
Pageins: 14072.
Pageouts: 0.
Object cache: 2791 hits of 6564 lookups (42% hit rate)

if it happens again, I'll post when it does start thrashing again.

 

[Bear]

...
Pageouts: 0.
...

You're not paging. Not with those stats. The swapfile is not getting used at all. You should just have one swapfile in /var/vm.

I average around 5 swapfiles on a 2GB PM G4. That's a lot of swapfile usage.

If your system starts acting up again capture the same data and also an "Ls -l /var/vm" for reference.

It might be 10.4.2 or the last security update. I've noticed a bit of oddness as well. I'm hoping 10.4.3 fixes it.

 

[superbovine]

You're not paging. Not with those stats. The swapfile is not getting used at all. You should just have one swapfile in /var/vm.

I average around 5 swapfiles on a 2GB PM G4. That's a lot of swapfile usage.

If your system starts acting up again capture the same data and also an "Ls -l /var/vm" for reference.

It might be 10.4.2 or the last security update. I've noticed a bit of oddness as well. I'm hoping 10.4.3 fixes it.

that was my baseline of 0

 

[Bear]

that was my baseline of 0

You should collect baselines every so often and then post the last one before it happens and then what it looks like after the thrashing happens.

 

[superbovine]

Mach Virtual Memory Statistics: (page size of 4096 bytes)
Pages free: 18118.
Pages active: 100376.
Pages inactive: 54413.
Pages wired down: 23698.
"Translation faults": 3338970.
Pages copy-on-write: 117541.
Pages zero filled: 1952930.
Pages reactivated: 81688.
Pageins: 35815.
Pageouts: 1772.


I keep getting "stuck" processes. apparently the lag is caused by the processes waiting on other processes to come out of deadlock. I believe they are waiting on the hard drive I/O.

FYI: dnetc is something like a folding program. it suppose to suck all the cpu speed and never giving me a problem. although there was just an update. I'll trying running without it for awhile. i still think i have a hard drive problem.

Processes: 68 total, 3 running, 2 stuck, 63 sleeping... 232 threads 00:05:54
Load Avg: 1.57, 1.59, 1.66 CPU usage: 82.3% user, 17.7% sys, 0.0% idle
SharedLibs: num = 188, resident = 37.1M code, 4.03M data, 6.12M LinkEdit
MemRegions: num = 9232, resident = 186M + 15.8M private, 125M shared
PhysMem: 91.3M wired, 342M active, 220M inactive, 654M used, 113M free
VM: 6.33G + 126M 35969(0) pageins, 1772(0) pageouts

PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE VSIZE
376 top 15.0% 0:23.08 1 18 22 720K 400K 1.15M 26.9M
375 mdimport 0.0% 0:00.65 4 64 56 1.10M 2.17M 3.17M 39.5M
373 Safari 0.5% 0:10.68 6 129 310 23.4M- 27.3M 35.4M- 185M-
367 bash 0.0% 0:00.03 1 14 17 208K 764K 876K 27.1M
366 login 0.0% 0:00.02 1 16 36 144K 416K 576K 26.9M
348 Mail 0.0% 0:03.99 7 142 271 6.23M 14.6M 15.9M 167M
289 Alerts Dae 0.0% 0:02.68 3 185 149 2.26M 6.93M 5.84M 154M
251 Skype 4.0% 1:58.57 13 262 408 24.1M 16.5M 33.6M 250M
243 iChatAgent 0.0% 0:01.52 5 84 131 2.25M 3.94M 5.45M 107M
242 iChat 0.0% 0:23.39 9 252 379 6.64M 17.4M 16.9M 187M
235 Microsoft 0.0% 0:23.11 2 83 144 6.74M 6.15M 9.85M 138M
234 Microsoft 0.9% 2:50.33 6 185 347 21.2M 27.2M 17.8M 212M
224 DashboardC 0.0% 0:13.59 3 79 152 4.20M 4.13M 6.77M 140M
223 DashboardC 0.0% 0:13.19 3 95 138 3.67M 4.13M 6.35M 139M
222 DashboardC 0.0% 0:13.96 3 79 149 4.25M 4.13M 6.77M 140M
221 DashboardC 0.0% 0:00.89 3 79 122 2.75M 6.36M 5.75M 147M
219 DashboardC 0.0% 0:00.63 3 95 111 2.89M 3.97M 5.40M 138M
218 DashboardC 0.0% 0:08.49 4 108 170 7.02M 5.56M 11.4M 151M
217 DashboardC 0.0% 0:01.17 3 79 128 3.26M 12.2M 6.18M 139M
216 DashboardC 0.0% 0:09.24 4 104 139 7.04M 4.11M 9.64M 141M
215 DashboardC 0.0% 0:01.31 3 100 183 4.80M 8.23M 9.01M 156M
214 DashboardC 4.5% 0:20.24 4 108 161 5.96M 6.38M 10.4M 151M
208 bash 0.0% 0:00.04 1 14 16 212K 764K 852K 27.1M
207 login 0.0% 0:00.02 1 16 36 132K 416K 564K 26.9M
205 AppleSpell 0.0% 0:00.60 1 33 34 696K 1.68M 2.11M 37.7M
198 dnetc 65.0% 3:54:25 2 18 23 228K 604K 828K 27.6M

 

[superbovine]

yeah i was right it was the hard drive. i get the flashing system icon now when i try to turn on my powerbook. i have to order combo drive and hard drive. my combo drive stop working months ago.