Results 1 to 10 of 10
  1. #1

    Server problem(apache?)

    Hello,

    I have this strange problem with my server, average bandwith is around 25Mpbs but sometimes it drops in 1 instance to 5Mbps and the site go very slow. And then after some hours without changing anything it goes back to normal, the sites load fast again.
    Its a P4 2.8Ghz with 1024MB ram. Its running a big site with 50K+ uniques visitors/day, its a php/mysql site(small queries).

    I also mailed the datacenter they say my server is connected via 1Gbps line to coreswitch so the network could not be the problem(they say)
    Normally I have 350 processen en server load is around 1, cpu 7%.Example site on the server http://domainshuffle.com/ (maybe fast again when you view it.)
    I tried restarting apache,mysql,whole server, cleaning tmp, looking for cron, did not changed anything.
    I think its a problem with apache because when I want to view server-status it sometimes will not show(see output below).

    I would really appreciate it if someone could help me.
    Thx in advance


    I will show some output when site are slow (at this moment).

    Code:
    top:
    top - 20:45:24 up 5 days, 35 min,  2 users,  load average: 0.36, 0.31, 0.33
    Tasks: 167 total,   1 running, 137 sleeping,   0 stopped,  29 zombie
    Cpu(s):  3.0% us,  4.2% sy,  0.0% ni, 88.5% id,  2.8% wa,  0.0% hi,  1.5% si
    Mem:   1034492k total,   996636k used,    37856k free,    21292k buffers
    Swap:  2096472k total,    61028k used,  2035444k free,   819480k cached
    
      PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
    18060 root      25   0 22868  13m 9792 S  2.3  1.3   2:45.60 httpd
     8952 root      16   0  8648 2216 1792 S  0.3  0.2   0:01.22 sshd
    15724 root      16   0  3736  992  748 R  0.3  0.1   0:00.01 top
    15816 apache    25   0     0    0    0 Z  0.3  0.0   0:00.01 httpd <defunct>
    15817 apache    23   0 23028  13m 9956 S  0.3  1.3   0:00.01 httpd
    15823 apache    25   0     0    0    0 Z  0.3  0.0   0:00.01 httpd <defunct>
        1 root      16   0  2644  452  420 S  0.0  0.0   0:00.80 init
        2 root      RT   0     0    0    0 S  0.0  0.0   0:00.35 migration/0
        3 root      34  19     0    0    0 S  0.0  0.0   0:00.88 ksoftirqd/0
        4 root      RT   0     0    0    0 S  0.0  0.0   0:00.36 migration/1
        5 root      34  19     0    0    0 S  0.0  0.0   0:01.15 ksoftirqd/1
        6 root       5 -10     0    0    0 S  0.0  0.0   0:00.00 events/0
        7 root       5 -10     0    0    0 S  0.0  0.0   0:00.00 events/1
        8 root       6 -10     0    0    0 S  0.0  0.0   0:00.00 khelper
        9 root      15 -10     0    0    0 S  0.0  0.0   0:00.00 kacpid
       29 root       5 -10     0    0    0 S  0.0  0.0   0:00.00 kblockd/0
    
    
    free -m
                      total       used       free     shared    buffers     cached
    Mem:          1010        986         24          0         20        814
    -/+ buffers/cache:        151        859
    Swap:         2047         59       1987
    
    
    
    Sometimes when I want to view server-status I get this:
    
    Looking up localhost
    Making HTTP connection to localhost
    Alert!: Unable to connect to remote host.
    lynx: Can't access startfile http://localhost/server-status
    
    
    And sometimes:
    
     Server uptime: 5 minutes 47 seconds
       Total accesses: 14 - Total Traffic: 411 kB
       CPU Usage: u.34 s.04 cu0 cs0 - .11% CPU load
       .0403 requests/sec - 1212 B/second - 29.4 kB/request
    
    
    hdparm -tT /dev/sda
    
    /dev/sda:
     Timing cached reads:   3080 MB in  2.00 seconds = 1538.70 MB/sec
    HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
     Timing buffered disk reads:  148 MB in  3.00 seconds =  49.32 MB/sec
    HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
    
    
    httpd.conf: (I have run along time with this config dont think this it the problem)
    
    Timeout 100
    KeepAlive On
    MaxKeepAliveRequests 150
    KeepAliveTimeout 1
    MinSpareServers 10
    MaxSpareServers 22
    StartServers 8
    MaxClients 550
    MaxRequestsPerChild 400
    Last edited by Dmitry85; 10-19-2005 at 04:05 PM.

  2. #2
    Join Date
    Apr 2003
    Location
    NC
    Posts
    3,093
    Everything above looks just fine. If you are having trouble when the bandwidth drops perhaps it is the provider? I would suggest setting up something like hotsanic that can graph the load and cpu, see if you can get any coorliation. From the above it looks like this may be an isp related problem since the disks are fast, loads low, and apache is doing almost nothing.
    John W, CISSP, C|EH
    MS Information Security and Assurance
    ITEagleEye.com - Server Administration and Security
    Yawig.com - Managed VPS and Dedicated Servers with VIP Service

  3. #3
    Thx, was thinking of that also(problem with isp) but when I try to look at server-status, it will not show up sometimes and thats local(like you see on output).
    Here is a testfile http://83.149.104.34/10mb.bin

  4. #4
    Join Date
    Mar 2002
    Location
    Toronto
    Posts
    237
    I'm having this same problem right now I think.
    Brand new box pretty much - and its for a very active site with around 200 and up people viewing hte site.

    Past few days was fine - I edited httpd.conf when I set the box up to increase the maxclients..Was running fine with anywhere between 260 to 380 processes at a time.

    Now it cant get above 125 - and which is basic. The apache processes running are about 5 or so at the most at a time - and keep getting killed (defunct).

    Apache Status in WHM says logging a lot - L - and I've tried everything I can think of - but I cant get it going again. Site is dragging and loading slowly for everyone.
    Any ideas?

  5. #5
    "Good" that I am not alone, when I look at top and press "c" to see the command I see this,
    "18060 root 25 0 22868 13m 9792 S 2.3 1.3 5:18.11 /usr/sbin/httpd -DHAVE_AUTOINDEX -DHAVE_CERN_META -DHAVE_REWRITE -DHAVE_AUTH -DHAVE_STATUS -DHAVE" on top.
    Dont know if it gives you any information.

  6. #6
    I have gone to sleep and server is backup running fine again, sites load fast. So I know what to do just go to sleep :p

    Anyone an idea how to solve this ?

  7. #7
    Join Date
    Mar 2002
    Location
    Toronto
    Posts
    237
    Hey Dmitry
    I solved my problem by clearing the apache error log - it was huge - AND the domlog for the site - again it was huge.
    If you go into ssh and type

    tail -150 /usr/local/apache/logs/error_log - or wherever your error log is and see a lot of error about file size exceeded - then odds are its your domlogs etc.,

    If so - you can go into WHM and check the option to delete the logs after stats run so that it doesnt get too big.

    Then do a regular (cron) logrotate on your apache access and error logs cuz they are bound to get big as well with a busy site like that.

    Hope this helps.

  8. #8
    Thanks but my logs are not that big, awstats cleans it.
    Biggest log is 534MB, and my logs are at /var/log/httpd/domains
    But when the problem comes back I will look at the logs if they are to big.

    Anyone any other suggestions?

  9. #9
    And again I am having the same problem, sites are very slow at the moment, anyone an idea what this?

    Thx in advance

  10. #10
    Found it, logs where to big 2GB, yihaaaaa!

    Thx ALOT CeciliaH !!
    Last edited by Dmitry85; 11-03-2005 at 07:28 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •