Re: Memory problem on Linux 2.6.5

On Thursday, May 13, 2004, at 06:06 PM, Theo Schlossnagle wrote:

> On May 13, 2004, at 8:16 PM, Clint Byrum wrote:
>> mod_backhand is showing erroneous results for the amount of memory on=20=

>> a machine when running Linux 2.6.5.
>>
>> Note:
>> Ent	Hostname	Age	Address		Total Mem		=
Avail Mem
>> --- --------- --- -------------  ------------  -----
>> 7	node2	1	1.2.3.5:80	1008=A0MB		850=A0MB=09=

>> 11	node3	1	1.2.3.4:80	4294967295 MB	4294967295
>>
>>
>> node 2 and node3 are identical hardware.. but node3 is running 2.6.5,=20=

>> and node2 is running 2.4.25. I'm sure this is due to the change in=20
>> /proc/meminfo.
>
> Rainer has sent me to patches that should have addressed this problem.=20=

>  I don't have access to a 2.6.x machine on which I can test=20
> mod_backhand at the moment.  Are you running the latest mod_backhand=20=

> from CVS?
>

No.. is there a stable branch that I can trust? And will running CVS in=20=

any way mess up my other servers that are not upgraded yet?

>> On a side note, Arriba also comes out *way* lower on this box running=20=

>> 2.6.5. Maybe the box is just slower running 2.6.5 ... or maybe its=20
>> the HyperThreading again. :-P
>
> Perhaps... you can manually set the arriba by editing the file it=20
> writes.  The arriba is a very arbitrary metric, so candidacy functions=20=

> and people should take it with a grain of salt.
>

Yes definitely. Though I think byLoad might take it too seriously...=20
because actual traffic across multiple identical machines is directly=20
proportional to Arriba, even if its only slightly different. I'm not=20
worried about it... we've started setting arriba manually on identical=20=

machines, and it has worked out well.=