[Issue] ProFTPD & Apache are slow

27 posts / 0 new
Last post
#1 Sat, 03/19/2011 - 04:08
DarkenedCell

[Issue] ProFTPD & Apache are slow

Hello there,

After a wile Virtualmin was running perfect on Ubuntu 10.04, for some reason it started giving error 500 on random periods and loading takes too long. Some times I am loading pages between 10 to 30 minutes with is not healthy. After rebooting it is temporary functioning properly but for some reason after 30 minutes up time it is loading slow.

I have the datacenter looked up on the switches and routes and its all okay since other services (TeamSpeak 3) and Ruby on Rails servers are loading insanely fast.

The error logs:

tail -f /var/log/apache2/error.log

PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Thu Mar 17 19:48:53 2011] [error] [client 109.230.221.171] File does not exist: /htdocs [Thu Mar 17 20:18:26 2011] [error] [client 109.230.221.171] File does not exist: /htdocs [Fri Mar 18 01:40:34 2011] [error] [client 88.80.10.1] File does not exist: /var/www/pp [Fri Mar 18 15:07:30 2011] [notice] mod_fcgid: call /home/baangracht/public_html/index.php with wrapper /home/baangracht/fcgi-bin/php5.fcgi PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0
Warning: Directive 'register_long_arrays' is deprecated in PHP 5.3 and greater in Unknown on line 0

Warning: Directive 'magic_quotes_gpc' is deprecated in PHP 5.3 and greater in Unknown on line 0

netstat:

netstat

Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 dedi.uxw.nl:ssh 546AC427.cm-12-3d:49238 ESTABLISHED tcp 0 0 dedi.uxw.nl:49353 clanserver4u.de.qu:ircd ESTABLISHED tcp6 0 0 2a00:dd0:0:14:20c:41395 denis.it.su.se:ircd ESTABLISHED udp 0 0 dedi.uxw.nl:52892 hardy.teamspeak.4p:2010 ESTABLISHED udp6 0 0 localhost:33976 localhost:33976 ESTABLISHED

At this moment it is running nothing iregular. Processes: SBNC (IRC Bouncer) TS3 Server (512 Slots) Virtualmin & Processes

Notes: CSS Loads fast, Binaries like: Images (Any kind) loads superb slow and generates the error 500 (random times)

Sat, 03/19/2011 - 04:17
DarkenedCell

I have to add this comment: It is slowing down on binary files (jpg/png/gif etc)

Sat, 03/19/2011 - 05:29 (Reply to #2)
DarkenedCell

Problem soulved: Had to install additional plugin with following command: apt-get install imagemagick

Dunno how I forgot to install this package*

Sat, 03/19/2011 - 09:27
andreychek

Great, glad you got it figured out! That a simpler fix than some of the things I was getting ready to suggest :-)

-Eric

Sat, 03/19/2011 - 15:59
DarkenedCell

Well, usually the Apache2 Package installs or suggests the Imagemagic libraries on Ubuntu packages from my knowledge. Simply Virtualmin skips those suggestions and heads on with a clean install.

Would it not be a suggestion for installing Imagemagic along with Apache?

Sat, 03/19/2011 - 16:13
andreychek

Well, reviewing the Apache2 packages over at packages.ubuntu.com, it doesn't look like imagemagick is listed as a suggestion for apache2, apache2-mpm-prefork, apache2.2-bin, or apache2.2-common.

However, even if it were -- for every person who wants to see each suggested package installed, there's another who doesn't want it installed due to wasted space or security concerns.

So in most cases, unless it's required by Virtualmin or the packages installed along with it -- it's generally up to the sysadmins to install the dependencies for the web apps they're looking to use :-)

-Eric

Sun, 03/20/2011 - 04:19
DarkenedCell

That clears things up, thank you :)

Sun, 03/20/2011 - 15:45
DarkenedCell

The same issue is back after a day, even I have restarted Apache.

Error log:

tail -f /var/log/apache2/error.log

[Sun Mar 20 21:43:43 2011] [notice] caught SIGTERM, shutting down [Sun Mar 20 21:43:44 2011] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec) [Sun Mar 20 21:43:44 2011] [notice] Digest: generating secret for digest authentication ... [Sun Mar 20 21:43:44 2011] [notice] Digest: done PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/apache2/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Sun Mar 20 21:43:45 2011] [notice] Apache/2.2.14 (Ubuntu) DAV/2 SVN/1.6.6 mod_fcgid/2.3.4 PHP/5.3.2-1ubuntu4.7 with Suhosin-Patch mod_ruby/1.2.6 Ruby/1.8.7(2010-01-10) mod_ssl/2.2.14 OpenSSL/0.9.8k configured -- resuming normal operations [Sun Mar 20 21:43:49 2011] [notice] mod_fcgid: call /home/pulse207/domains/test.pulse207.com/public_html/index.php with wrapper /home/pulse207/domains/test.pulse207.com/fcgi-bin/php5.fcgi PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Sun Mar 20 21:43:58 2011] [notice] mod_fcgid: call /home/pulse207/public_html/index.php with wrapper /home/pulse207/fcgi-bin/php5.fcgi PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0

from other sites there are no entries tho*

Sun, 03/20/2011 - 15:59
andreychek

Howdy,

I don't actually see any errors or warnings in your output that would suggest why things are working slowly.

If you run the command "uptime", what does it show?

And if you run "top", are there any programs at the top of that output that it's showing are CPU hungry?

-Eric

Sun, 03/20/2011 - 16:31
DarkenedCell

Info from TOP: top - 22:28:00 up 4 days, 12:03, 1 user, load average: 0.49, 0.40, 0.33 Tasks: 172 total, 1 running, 171 sleeping, 0 stopped, 0 zombie Cpu(s): 0.3%us, 0.1%sy, 0.0%ni, 99.5%id, 0.0%wa, 0.0%hi, 0.1%si, 0.0%st Mem: 4060212k total, 1735008k used, 2325204k free, 184652k buffers Swap: 94200k total, 0k used, 94200k free, 748912k cached

See attachment for more info, and also heading for bed since I got to head to work tomorrow.

Sun, 03/20/2011 - 17:31
andreychek

Hmm, you're load isn't particularly high. I don't think that's the problem.

What if you change it to use CGI rather than FCGID -- does that help? You can change that by going into Services -> Website Options, and change the PHP Execution Mode.

-Eric

Mon, 03/21/2011 - 02:37
DarkenedCell

Changed it to CGI wrapper now througgt my mobile, lets hope it works I will inform once I am at my work*

Mon, 03/21/2011 - 03:43
DarkenedCell

Recent Error Messages:

[Mon Mar 21 09:28:45 2011] [notice] Digest: generating secret for digest authentication ... [Mon Mar 21 09:28:45 2011] [notice] Digest: done PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/apache2/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Mon Mar 21 09:28:46 2011] [notice] Apache/2.2.14 (Ubuntu) DAV/2 SVN/1.6.6 mod_fcgid/2.3.4 PHP/5.3.2-1ubuntu4.7 with Suhosin-Patch mod_ruby/1.2.6 Ruby/1.8.7(2010-01-10) mod_ssl/2.2.14 OpenSSL/0.9.8k configured -- resuming normal operations [Mon Mar 21 09:30:58 2011] [notice] caught SIGTERM, shutting down [Mon Mar 21 09:30:59 2011] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec) [Mon Mar 21 09:30:59 2011] [notice] Digest: generating secret for digest authentication ... [Mon Mar 21 09:30:59 2011] [notice] Digest: done PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/apache2/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Mon Mar 21 09:30:59 2011] [notice] Apache/2.2.14 (Ubuntu) DAV/2 SVN/1.6.6 mod_fcgid/2.3.4 PHP/5.3.2-1ubuntu4.7 with Suhosin-Patch mod_ruby/1.2.6 Ruby/1.8.7(2010-01-10) mod_ssl/2.2.14 OpenSSL/0.9.8k configured -- resuming normal operations

Problem is still there : /

Mon, 03/21/2011 - 09:19
andreychek

Howdy,

Unfortunately, I'm still not seeing any related error messages in your Apache output. From that log, everything appears normal.

You would want to look at the Apache log specific to the domain that's having problems though... that would be in $HOME/logs/error_log.

Now, is it still both Apache and ProFTPd that are slow?

If that's the case, what you may be seeing is a network problem of some sort. One place to start would be to make sure that you aren't experiencing any packet loss between your desktop and your server. You could ping your server, let it run for a minute or two, and make sure all the packets arrive.

If you're using a VPS, it's possible that another VPS user on your same host is using a lot of disk IO, making everything on your VPS appear slow. If that's the case, you may need to ask your provider about that particular situation.

-Eric

Mon, 03/21/2011 - 09:28
DarkenedCell

The logs at the home directories are not giving more detail it is all about missing files or that mycrypt error messages as posted below.

[Mon Mar 21 14:18:47 2011] [error] [client 145.7.224.70] PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0 [Mon Mar 21 14:19:58 2011] [error] [client 145.7.224.104] PHP Deprecated: Comments starting with '#' are deprecated in /etc/php5/cgi/conf.d/mcrypt.ini on line 1 in Unknown on line 0, referer: http://www.pulse207.com/

The server is running on a personal Dedicated server with means its 100% used by me, any idea how to check the disk activity?

Mon, 03/21/2011 - 09:30
DarkenedCell

Attaching the ping log:

Mon, 03/21/2011 - 09:49
DarkenedCell

After doing some research I found the tool IOTOP to check disk activity, its some times writing some thing to log files with is maximum up to 53.24B/s

Further no activity at all see attachment for screenshot.

Mon, 03/21/2011 - 09:51
andreychek

The messages you're seeing in your error logs are just mentioning that, in newer PHP versions, comments will need to be made with a ; character rather than a # character. They're just notices, and wouldn't explain why things are slow.

As far as your ping output goes -- the Windows ping is apparently not all that helpful :-)

We need to see the packet number that's being sent out, to make sure each one is arriving, and none are being skipped.

I don't have a Windows box in front of me to experiment with, but you could try passing in a "-c 50" to ping... I think that'll tell it to send 50 packets, and it should then tell you if all were received back.

As far as disk IO is concerned -- if that's a dedicated server, and it's yours -- I wouldn't worry about that. If you had high disk IO, it would show up in your load average when you ran "uptime". In your case, your load is pretty low.

-Eric

Mon, 03/21/2011 - 10:03
DarkenedCell

Directly from the server:

# ping -c 50 www.google.nl
PING www.l.google.com (66.102.13.103) 56(84) bytes of data.
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=1 ttl=57 time=14.8 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=2 ttl=57 time=5.83 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=3 ttl=57 time=31.7 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=4 ttl=57 time=23.3 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=5 ttl=57 time=5.37 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=6 ttl=57 time=5.79 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=7 ttl=57 time=5.84 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=8 ttl=57 time=5.65 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=9 ttl=57 time=5.55 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=10 ttl=57 time=5.53 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=11 ttl=57 time=5.89 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=12 ttl=57 time=5.57 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=13 ttl=57 time=5.39 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=14 ttl=57 time=5.76 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=15 ttl=57 time=5.63 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=16 ttl=57 time=5.26 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=17 ttl=57 time=5.70 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=18 ttl=57 time=5.38 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=19 ttl=57 time=5.54 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=20 ttl=57 time=5.96 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=21 ttl=57 time=5.58 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=22 ttl=57 time=5.70 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=23 ttl=57 time=5.37 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=24 ttl=57 time=5.56 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=25 ttl=57 time=5.51 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=26 ttl=57 time=5.56 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=27 ttl=57 time=9.17 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=28 ttl=57 time=5.37 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=29 ttl=57 time=5.31 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=30 ttl=57 time=5.37 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=31 ttl=57 time=5.56 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=32 ttl=57 time=5.72 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=33 ttl=57 time=5.55 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=34 ttl=57 time=5.52 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=35 ttl=57 time=5.51 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=36 ttl=57 time=5.26 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=37 ttl=57 time=5.31 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=38 ttl=57 time=5.35 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=39 ttl=57 time=5.71 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=40 ttl=57 time=5.75 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=41 ttl=57 time=5.34 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=42 ttl=57 time=5.51 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=43 ttl=57 time=5.81 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=45 ttl=57 time=5.54 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=46 ttl=57 time=5.65 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=47 ttl=57 time=5.40 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=48 ttl=57 time=5.35 ms
64 bytes from ez-in-f103.1e100.net (66.102.13.103): icmp_seq=49 ttl=57 time=5.28 ms

--- www.l.google.com ping statistics ---
50 packets transmitted, 48 received, 4% packet loss, time 49078ms
rtt min/avg/max/mdev = 5.262/6.739/31.792/4.655 ms

Mon, 03/21/2011 - 10:11
DarkenedCell

Directly from server to a local DNS Profider:

# ping -c 50 www.transip.nl
PING www.transip.nl (80.69.67.15) 56(84) bytes of data.
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=1 ttl=60 time=1.32 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=2 ttl=60 time=1.38 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=3 ttl=60 time=1.32 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=4 ttl=60 time=1.32 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=5 ttl=60 time=1.21 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=6 ttl=60 time=1.31 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=7 ttl=60 time=1.27 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=8 ttl=60 time=1.29 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=9 ttl=60 time=1.35 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=10 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=11 ttl=60 time=1.27 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=12 ttl=60 time=1.23 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=13 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=14 ttl=60 time=1.33 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=15 ttl=60 time=13.2 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=16 ttl=60 time=1.33 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=17 ttl=60 time=1.30 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=18 ttl=60 time=1.29 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=19 ttl=60 time=1.29 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=20 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=21 ttl=60 time=1.22 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=22 ttl=60 time=1.40 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=23 ttl=60 time=1.25 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=24 ttl=60 time=1.27 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=25 ttl=60 time=1.28 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=26 ttl=60 time=1.24 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=27 ttl=60 time=1.22 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=28 ttl=60 time=1.34 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=29 ttl=60 time=1.33 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=30 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=31 ttl=60 time=1.24 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=32 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=33 ttl=60 time=1.27 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=34 ttl=60 time=1.36 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=35 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=36 ttl=60 time=1.42 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=37 ttl=60 time=5.17 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=38 ttl=60 time=1.33 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=39 ttl=60 time=1.41 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=40 ttl=60 time=1.39 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=41 ttl=60 time=2.30 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=42 ttl=60 time=1.27 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=43 ttl=60 time=1.23 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=44 ttl=60 time=1.24 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=45 ttl=60 time=3.51 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=46 ttl=60 time=1.49 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=47 ttl=60 time=1.26 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=48 ttl=60 time=1.25 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=49 ttl=60 time=1.23 ms
64 bytes from www.transip.nl (80.69.67.15): icmp_seq=50 ttl=60 time=1.40 ms

--- www.transip.nl ping statistics ---
50 packets transmitted, 50 received, 0% packet loss, time 49059ms
rtt min/avg/max/mdev = 1.218/1.683/13.221/1.765 ms

Mon, 03/21/2011 - 10:12
andreychek

Yeah, it looks like it skipped a couple of packets in that output.

So you may indeed be seeing a network problem of some sort. If you send 50 packets to your server, you should be able to rely on the fact that all will be returned. If that doesn't happen, something between your desktop and your server is wrong.

This is where it gets tough -- it could be anything :-)

It could be a network card, network cable, router, switch, hub, and so on.

You can use tools such as traceroute to help you diagnose the issue -- it can help you figure out roughly where the problem is occurring. I like the program "mtr" for these sorts of issues, though it's only available for Linux.

-Eric

Mon, 03/21/2011 - 10:15
DarkenedCell

I will supply some more pings during this day and contact my datacenter in order to check the hardware since this is going on for month's already. And they did checked it a couple of times so far...

I am on IRC also by the way under the name "Companion"

Mon, 03/21/2011 - 10:17
andreychek

It doesn't look like you are always getting packet loss. It shows up in some cases, and not others. If you only ever see it when doing the ping from your desktop, that may mean it's something with your desktop machine, or perhaps your home ISP.

Or, it could just mean the problem is intermittent :-)

But, the above output definitely shows a problem.

-Eric

Mon, 03/21/2011 - 10:23 (Reply to #23)
DarkenedCell

It happened on multiple ISP's. Have this issue from: KPN, Ziggo, directadsl, XS4All and UPC

Mon, 03/21/2011 - 10:21
DarkenedCell

Ran MTR a couple of times and these are the results:

# mtr -r -c 1 www.transip.nl
HOST: dedi Loss% Snt Last Avg Best Wrst StDev
1. 2a00:dd0:0:14::1 0.0% 1 1.2 1.2 1.2 1.2 0.0
2. 2a00:dd0::1:13 0.0% 1 0.3 0.3 0.3 0.3 0.0
3. 2a00:dd0::1:5 0.0% 1 0.4 0.4 0.4 0.4 0.0
4. 2a00:dd0::1:2 0.0% 1 0.9 0.9 0.9 0.9 0.0
5. amsix.router1.dcg.transip.ne 0.0% 1 1.5 1.5 1.5 1.5 0.0
6. www.transip.nl 0.0% 1 1.7 1.7 1.7 1.7 0.0
# mtr -r -c 1 www.google.nl
HOST: dedi Loss% Snt Last Avg Best Wrst StDev
1. 94.228.217.65 0.0% 1 4.5 4.5 4.5 4.5 0.0
2. r3.alp.nl.netrouting.net 0.0% 1 0.5 0.5 0.5 0.5 0.0
3. r1.ams1.nl.netrouting.net 0.0% 1 1.1 1.1 1.1 1.1 0.0
4. core1.ams.net.google.com 0.0% 1 1.3 1.3 1.3 1.3 0.0
5. 209.85.248.93 0.0% 1 1.6 1.6 1.6 1.6 0.0
6. 216.239.49.38 0.0% 1 6.4 6.4 6.4 6.4 0.0
7. 64.233.174.133 0.0% 1 16.0 16.0 16.0 16.0 0.0
8. ez-in-f99.1e100.net 0.0% 1 5.6 5.6 5.6 5.6 0.0
# mtr -r -c 1 irc.freenode.net
HOST: dedi Loss% Snt Last Avg Best Wrst StDev
1. 2a00:dd0:0:14::1 0.0% 1 0.5 0.5 0.5 0.5 0.0
2. 2a00:dd0::1:13 0.0% 1 0.3 0.3 0.3 0.3 0.0
3. 2a00:dd0::1:5 0.0% 1 0.5 0.5 0.5 0.5 0.0
4. 2a00:dd0::1:2 0.0% 1 1.6 1.6 1.6 1.6 0.0
5. 20gigabitethernet1-3.core1.a 0.0% 1 1.5 1.5 1.5 1.5 0.0
6. 10gigabitethernet1-1.core1.f 0.0% 1 8.3 8.3 8.3 8.3 0.0
7. te-3-4.ar2.fra4.gblx.net 0.0% 1 8.7 8.7 8.7 8.7 0.0
8. 2001:450:2002:56::2 0.0% 1 21.0 21.0 21.0 21.0 0.0
9. 2001:1418:1:700::a 0.0% 1 20.8 20.8 20.8 20.8 0.0
10. calkins.freenode.net 0.0% 1 22.8 22.8 22.8 22.8 0.0
# mtr -r -c 1 www.google.com
HOST: dedi Loss% Snt Last Avg Best Wrst StDev
1. 94.228.217.65 0.0% 1 0.5 0.5 0.5 0.5 0.0
2. r3.alp.nl.netrouting.net 0.0% 1 0.3 0.3 0.3 0.3 0.0
3. r1.ams1.nl.netrouting.net 0.0% 1 0.9 0.9 0.9 0.9 0.0
4. core1.ams.net.google.com 0.0% 1 23.9 23.9 23.9 23.9 0.0
5. 209.85.248.93 0.0% 1 1.6 1.6 1.6 1.6 0.0
6. 216.239.49.38 0.0% 1 5.4 5.4 5.4 5.4 0.0
7. 64.233.174.133 0.0% 1 5.6 5.6 5.6 5.6 0.0
8. ez-in-f106.1e100.net 0.0% 1 5.9 5.9 5.9 5.9 0.0

Mon, 03/21/2011 - 10:29
andreychek

When you run mtr, you don't necessarily want to use a "-c 1" -- that's not quite enough data. You'll want to let it sit for awhile. If you allow it to run for a bit, are you seeing packet loss when going to the sites you tried above?

If your issue happened at multiple data centers, then your issue is likely either with your server or your desktop (or desktop ISP) :-)

One thing you might try is to run:

ifconfig eth0

With that, do you see any RX or TX errors?

-Eric

Mon, 03/21/2011 - 10:39 (Reply to #26)
DarkenedCell

It has always been at the same datacenter (this server), this stuff hapened when we re-installed the server cause of a corrupted kernel. Further its pure server side cause switches have been reset and rebooted from my knowledge. Users can only access the websites with are slowly loading. The ISP's I named were from the home users.

The sites that are currently run are: http://www.pulse207.com/ http://test.pulse207.com/ http://www.lianneoost.com/ http://www.michelsmind.com/ http://www.baangracht.nl/

CSS loads rapid but images are still a pain :(

Will forward this topic to the datacenter*