nightly 2.6 LXR run?

nightly 2.6 LXR run?

[Jon Masters]

Hi folks,

So I'm curious (before I do it) whether anyone is currently running an
automated nightly LXR against Linus' git kernel tree?

I know of coywolf's excellent resource, but since I'm playing with LXR
for some other stuff at the moment it'd be a good moment to get it done.

Jon.

Re: nightly 2.6 LXR run?

[Oleg Verych]

Hallo.
On 2006-11-14, Jon Masters wrote:
> Hi folks,
>
> So I'm curious (before I do it) whether anyone is currently running an
> automated nightly LXR against Linus' git kernel tree?

I don't think so. Also, i don't think, it's useful for development.
There are TAGS/tags make targets for it, and they're more productive, i think.

But what really will be helpful (well, for me), is another git-web public 
service on Linus' git tree. kernel.org is very heavy loaded with many
things, so normally i get "The load average on the server is too high",
when trying to trace some code history and reading logs. There were
some lkml posts, even from Linus, stating, that kernel.org is very
loaded and slow.

I don't want to load it much with git pull also, i just need history and
logs. Hope somebody will benefit from it too.

> I know of coywolf's excellent resource, but since I'm playing with LXR
> for some other stuff at the moment it'd be a good moment to get it done.
>
> Jon.
____

Re: nightly 2.6 LXR run?

[Auke Kok]

Oleg Verych wrote:
> Hallo.
> On 2006-11-14, Jon Masters wrote:
>> Hi folks,
>>
>> So I'm curious (before I do it) whether anyone is currently running an
>> automated nightly LXR against Linus' git kernel tree?
> 
> I don't think so. Also, i don't think, it's useful for development.
> There are TAGS/tags make targets for it, and they're more productive, i think.
> 
> But what really will be helpful (well, for me), is another git-web public 
> service on Linus' git tree. kernel.org is very heavy loaded with many
> things, so normally i get "The load average on the server is too high",
> when trying to trace some code history and reading logs. There were
> some lkml posts, even from Linus, stating, that kernel.org is very
> loaded and slow.

try to use `git2.kernel.org` instead of `git.kernel.org`. The second server somehow only 
has an average load of ~4 instead of ~250.

Cheers,

Auke

git web (Re: nightly 2.6 LXR run?)

[Oleg Verych]

On Tue, Nov 14, 2006 at 03:19:20PM -0800, Auke Kok wrote:
[]
> try to use `git2.kernel.org` instead of `git.kernel.org`. The second server 
> somehow only has an average load of ~4 instead of ~250.

Looking closer, i see all gitweb links (C) on both servers links are poining
to "www.kernel.org/git/?...". So, i thing there's no gitweb dup, only file
storage mirror.

BTW
,--
|olecom@flower:~$ host git.kernel.org
|git.kernel.org          CNAME   zeus-pub.kernel.org
|zeus-pub.kernel.org     A       204.152.191.37
|olecom@flower:~$ host git2.kernel.org
|git2.kernel.org         CNAME   zeus-pub2.kernel.org
|zeus-pub2.kernel.org    A       204.152.191.37
|olecom@flower:~$
`--
(maybe it's some way of multiplexing...)

And still "The load average on the server is too high".

:oE
____

Re: git web (Re: nightly 2.6 LXR run?)

[Auke Kok]

Oleg Verych wrote:
> On Tue, Nov 14, 2006 at 03:19:20PM -0800, Auke Kok wrote:
> []
>> try to use `git2.kernel.org` instead of `git.kernel.org`. The second server 
>> somehow only has an average load of ~4 instead of ~250.
> 
> Looking closer, i see all gitweb links (C) on both servers links are poining
> to "www.kernel.org/git/?...". So, i thing there's no gitweb dup, only file
> storage mirror.
> 
> BTW
> ,--
> |olecom@flower:~$ host git.kernel.org
> |git.kernel.org          CNAME   zeus-pub.kernel.org
> |zeus-pub.kernel.org     A       204.152.191.37
> |olecom@flower:~$ host git2.kernel.org
> |git2.kernel.org         CNAME   zeus-pub2.kernel.org
> |zeus-pub2.kernel.org    A       204.152.191.37
> |olecom@flower:~$
> `--
> (maybe it's some way of multiplexing...)
> 
> And still "The load average on the server is too high".

zeus1 is apparently offline due to a fan failure, so someone is playing mind tricks on 
you with DNS

Auke