Re: kernel 2.6.25-rc7 highly unstable on high load

view
thread

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

From: Eric Dumazet

Subject: Re: kernel 2.6.25-rc7 highly unstable on high load

Date: Thursday, March 27, 2008 - 9:07 am

Denys Fedoryshchenko a écrit :
quoted text> I did this tuning. Not sure what does mean rt_cache_entries, but seems they 
> are still increasing.
>
> Kup /config #  rtstat -c1000 -i60
> rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|
> rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|rt_cache|
> rt_cache|
>  entries|  in_hit|in_slow_|in_slow_|in_no_ro|  in_brd|in_marti|in_marti| 
> out_hit|out_slow|out_slow|gc_total|gc_ignor|gc_goal_|gc_dst_o|in_hlist|
> out_hlis|
>         |        |     tot|      mc|     ute|        |  an_dst|  
> an_src|        |    _tot|     _mc|        |      ed|    miss| verflow| 
> _search|t_search|
>   274234| 1590009|  271953|       0|    1809|       6|       0|       0|   
> 24551|    7365|       0|  279991|  279867|      24|       0| 5438630|   89656|
>   280535|   25463|   19956|       0|     187|       0|       0|       0|     
> 485|     263|       0|   20407|   20404|       3|       0|   12796|     242|
>   288072|   25367|   19830|       0|     163|       0|       0|       0|     
> 493|     256|       0|   20248|   20245|       3|       0|   17146|     346|
>   294943|   25689|   19807|       0|     154|       0|       0|       0|     
> 438|     250|       0|   20211|   20208|       3|       0|   21554|     432|
>   301118|   25287|   19645|       0|     170|       0|       0|       0|     
> 466|     236|       0|   20052|   20048|       3|       0|   25713|     528|
>   307651|   25663|   20264|       0|     180|       0|       0|       0|     
> 472|     234|       0|   20678|   20675|       3|       0|   30588|     607|
>   313767|   25810|   20014|       0|     175|       0|       0|       0|     
> 459|     237|       0|   20425|   20422|       3|       0|   33701|     699|
>
>
>   
Then try to lower gc_elasticity, to 3, or even 2

echo 3 >/proc/sys/net/ipv4/route/gc_elasticity






--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Thu Mar 27, 9:07 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Thu Mar 27, 9:29 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Thu Mar 27, 11:37 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Thu Mar 27, 11:56 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denis V. Lunev, (Thu Mar 27, 12:05 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, David Miller, (Thu Mar 27, 3:03 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Thu Mar 27, 5:47 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Thu Mar 27, 9:50 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Thu Mar 27, 10:25 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Thu Mar 27, 10:49 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Fri Mar 28, 12:38 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Fri Mar 28, 12:46 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Fri Mar 28, 2:50 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Fri Mar 28, 8:57 am)

Re: kernel 2.6.25-rc7 highly unstable on high load, Stephen Hemminger, (Fri Mar 28, 1:23 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Fri Mar 28, 1:45 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Eric Dumazet, (Fri Mar 28, 2:17 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, David Miller, (Fri Mar 28, 4:21 pm)

Re: kernel 2.6.25-rc7 highly unstable on high load, Denys Fedoryshchenko, (Sat Mar 29, 2:35 am)

Navigation

Popular discussions


linux-kernel:
Dave Jones	Re: OT: character encodings (was: Linux 2.6.20-rc4)
Greg Kroah-Hartman	[PATCH 17/36] sysdev: detect multiple driver registrations
Sam Ravnborg	Re: [PATCH] kbuild: fix make V=1
Nick Piggin	Re: [PATCH 0/24] make atomic_read() behave consistently across all architectures
Greg Kroah-Hartman	[PATCH 16/36] driver core: cpu: fix section mismatch in cpu.c:store_online
git:
Stephen R. van den Berg	Re: [RFC] origin link for cherry-pick and revert
Junio C Hamano	Re: [PATCH 1/2] Teach git-describe to display distances from tags.
Johannes Schindelin	Re: [PATCH 2/2] git-svn: support fetch with autocrlf on
fantasy1215	Will tortoisesvn conflict with tortoisegit?
Junio C Hamano	Re: [PATCH 6/6] Teach core object handling functions about gitlinks
linux-netdev:
Jarek Poplawski	Re: [PATCH] flush_work_sync vs. flush_scheduled_work Re: [PATCH] PHYLIB: IRQ event...
Lennert Buytenhek	Re: Distributed Switch Architecture(DSA)
Daniel Schaffrath	Re: tcp bw in 2.6
Guo-Fu Tseng	Re: jme: UDP checksum error, and lots of them
Gerrit Renker	[PATCH 37/37] dccp: Debugging functions for feature negotiation
openbsd-misc:
Claudio Jeker	Re: Vlan Tag on Vlan Tag (l2tunneling)
Josh Grosse	ssh/sshd challenge-response seems to have stopped working in -current
Pieter Verberne	File collision while using pkg_add
Tomas Bodzar	bsd: uvm_mapent_alloc: out of static map entries
Community First Financial	Teacher A+ Loan
git-commits-head:
Linux Kernel Mailing List	ath9k: Added get_survey callback in order to get channel noise
Linux Kernel Mailing List	tracing: protect reader of cmdline output
Linux Kernel Mailing List	kconfig: recalc symbol value before showing search results
Linux Kernel Mailing List	[ARM] 5185/1: Fix spi num_chipselect for lubbock
Linux Kernel Mailing List	swsusp: provide users with a hint about the no_console_suspend option

Colocation donated by:

Syndicate