Mailing Lists: Apple Mailing Lists

Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: web server woes!



From: Niels Meersschaert <email@hidden>

Lookupd is the process responsible for host lookups as others have
suggested.  Have you tried eliminating the dynamic sites (php &
mediawiki) & see if the problem continues?

That's a bit more trouble. So far, I haven't touched my httpd.conf (and included) files. Also, I'll have customers out of action long enough to see what is happening. Considering I can go a whole day without a problem, that's unacceptable at the moment.


It may be possible that
either the php app or mediawiki is trying to do a DNS lookup & thus
forcing lookupd to be so heavy.  A poorly coded php app that does DNS
lookups & doesn't cache the results would experience the behavior you
are describing.

I'll look into that with MediaWiki, although I believe I have all the right config options to NOT make it do lookups. Also, it is rarely a MediaWiki page at the end of log when things start going bad.


Also, you mentioned that you run your own DNS
server.  Is that on the same server, or a different machine?  Is this
machine running OS X Server or OS X Client?

Same machine, OS X Client. (As near as I can tell, the Server only buys me a fancy admin GUI for what I'm doing.)


From: Max Roeleveld - Qualion Internet Services <email@hidden>

When I have this problem, it isn't necessarily the slow MediaWiki
sites that are stuck. But it's always a spider at the end of the log.
(Of course, that may simply be statistical probability... :-)

Maybe, but it might also mean that a lot of requests kill of the server.

Just to clarify, nothing is actually "killed" (in Unix terms).

The latest case appears to be a DOS attack. I had some thirty requests for the same JPEG image within two minutes! I reported it to their ISP and blocked them in the router. There was also a virus infection attempt not long before: a SEARCH with a gazillion NUL characters. Reminds me of the awful "Code Red" days... AAARRRGGGHHH!!!

Depending on how much memory the G4 has...

1.12GB. It's been plenty so far.

Ah. MediaWiki can use memcached -- dunno if/how that runs on OS X, but I'd assume it would work.

When I first installed MediaWiki, I tried to install memcached, but it didn't go smoothly, and I didn't have time to figure out why it wouldn't compile.


Anyone know of a memchached that installs cleanly on MacOS X?

Here's what top -u looks like at the moment, when it's working fine:
PhysMem: 124M wired, 118M active, 349M inactive, 592M used, 559M
free
VM: 4.28G + 70.1M 69701(0) pageins, 236728(0) pageouts

Nothing too spectecular here -- you do have memory to burn, though, so
I'd really look into the memcached thing. Especially since your machine
might be underpowered.

G4/350 was fine until I installed MediaWiki... :-(

I would like to get memcached working, but I don't believe it would solve my current difficulties.

I'd really look into upping the
amount of spareservers for apache -- can't hurt.

Okay, I'll try it the next time I have to restart Apache. I don't want to subject my users to any unnecessary restarts, since I'm doing it several times a day anyway...


What does console.log say, if anything?

DOH! I should have known to go looking through /var/log/ for weird stuff...


netinfo.log has a lot of:

Jun 14 04:28:57 dns lookupd[5986]: NetInfo connection failed for server 127.0.0.1/local
Jun 14 04:37:24 dns last message repeated 6 times
Jun 14 04:50:57 dns last message repeated 5 times
Jun 14 04:56:55 dns last message repeated 3 times
Jun 14 05:07:52 dns last message repeated 5 times
Jun 14 05:20:56 dns last message repeated 2 times
Jun 14 05:29:55 dns last message repeated 2 times
Jun 14 05:40:55 dns last message repeated 12 times
...


(lookupd's PID is so high because I had tried restarting.)

Oh, and connected to what Walter Lee Davis meant (I think):
what is the ServerName configured?

It is different for each site. At the end of /etc/httpd.conf, I have: Include /private/etc/httpd/users/[^.]*.conf

and each file in users has a separate ServerName, which I believe is the proper way to do it. In any event, that part has been the same forever.

Is UseCanonicalName set to On or Off?

# fgrep UseCanonicalName /etc/httpd/httpd.conf
# UseCanonicalName: (new for 1.3) With this setting turned on, whenever
UseCanonicalName On


I suppose I'm kinda trying to say "Dude, maybe you should throw some
more CPU muscle at it". =]

Well, it's been working just fine until fairly recently!

I added MediaWiki in January, and it's run smoothly (well yea, a bit slow) until just a couple weeks ago.

Even more curious than before,

Me too! I'm going to try to figure out if those netinfo.log messages are meaningful.


I guess I should also stop and install 10.3.9. When things are running well, I HATE installing upgrades, because invariably, things break afterward... :-(

Thanks, all, for the great suggestions -- I wish some of them worked!

:::: If you succumb to the temptation of using violence in the struggle, unborn generations will be the recipients of a long and desolate night of bitterness, and your chief legacy to the future will be an endless reign of meaningless chaos. - Martin Luther King Jr.
:::: Jan Steinman <http://www.Bytesmiths.com/Item/99-0590-18>



_______________________________________________ Do not post admin requests to the list. They will be ignored. Web-dev mailing list (email@hidden) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/web-dev/email@hidden

This email sent to email@hidden


Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.