Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


My online.net server lagged and slow
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

My online.net server lagged and slow

recently, after some update, my server become sooooo slow. even login to ssh can takes for minutes.
how can I analyse this problem? I dont want to reinstall the server because I would lost all the files I kept..
here a screenshot for top command

«1

Comments

  • You have some serious load on your server. But it seems like the problem is not on your server, but on the node. You should create a support ticket and ask why the lode on the node is so high. Probably a user that abuse his server.

  • Give that webalizer a break!

  • Looks like you've used almost all memory, try cleaning memory cache, should probably solve the issue.

  • I noticed these 3 process takes high cpu usage. monitor.pl, miniserv.pl, webalizer..
    I am using virtualmin. I just scared to disable those processes because it may stop my panel from functioning well

  • Your problem is that the server has problems writing data to the disk (high I/O wait).

    Thanked by 3Mark_R netomx Noerman
  • @alexvolk said:
    Looks like you've used almost all memory, try cleaning memory cache, should probably solve the issue.

    I even reboot the server for many times.. but this problem will appear again..
    I just host small website on it with transmission seedbox.. It works well for previous month. I noticed this problem after some updates I made.. Do I need to open a ticket or I can solve this by reinstall the server?

  • @skagerrak said:
    Your problem is that the server has problems writing data to the disk (high I/O wait).

    how can I solve this??

  • xyzxyz Member

    Try running iotop

  • @psycholyzern said:
    how can I solve this??

    Have a look at the processes and why they have such problems. Do a ps auxf. Have a look at each of those processes. Have a look at their logs. If it's a dedicated server this might also indicate that your disk is failing, have a look at it using smartctl.

  • @xyz I dont have that installed on my centos. tried to install but the yum process seems lagged and stucked after I run yum install

  • Is this the Atom server?

  • @ErawanArifNugroho said:
    Is this the Atom server?

    yah.. the 1TB disk dedi

  • @skagerrak said:
    Have a look at the processes and why they have such problems. Do a ps auxf. Have a look at each of those processes. Have a look at their logs. If it's a dedicated server this might also indicate that your disk is failing, have a look at it using smartctl.

    seems like too hard for me.. If I reinstall the server, would this problem goes away?

  • Due to the slow cpu and also the disk io, the load might stay at high. When I use the Atom server, it's not good enough, so we need to change to the better plan.

    Disk io at the Atom server is not good.

  • @ErawanArifNugroho said:
    Due to the slow cpu and also the disk io, the load might stay at high. When I use the Atom server, it's not good enough, so we need to change to the better plan.

    Disk io at the Atom server is not good.

    It works well for previous 2 month.. I use it for same purpose, but this problem appears recently..

  • @psycholyzern do these Atoms use hard disks or iSCSI disks?

  • maybe @ErawanArifNugroho can help me answer this because he is using dedi from online.net too..is there anyway to check it?? I cant tell you because I didnt know how to recognise them

  • I think I would go fresh installing to restore default server programs.. maybe I did some changes that I dont realise.. is it ok for me??

  • @psycholyzern said:
    I think I would go fresh installing to restore default server programs.. maybe I did some changes that I dont realise.. is it ok for me??

    You could give it a try, if it's a disk issue as other's have suggested your going to lose the data anyway if it needs to be replaced.

    If a reinstall doesn't work then you're going to have to open a ticket and likely have the server or drive replaced.

  • @DigitalDuke said:
    If a reinstall doesn't work then you're going to have to open a ticket and likely have the server or drive replaced.

    then, I will go for reinstall.. I will keep this thread updated with my issue.. thanks for all

  • The lest option, we can try to reinstall it. But if the problem still occur, try to stop the webmin.

  • @psycholyzern good luck getting this fixed, I've used online.net for a few months, but never needed support, so I've no idea how helpful they are.

  • Before re-installing I would highly suggest you use smartctl to check the status of your hard drive. If it is having a lot of failures this could explain the extremely slow IO speed. This article should give you the generics in doing this check: Click Here

    If you can confirm its not the drive starting to die, then I would make sure to review your system message log (dmesg) and confirm you see no other hardware giving errors, such as the SATA controller on the server.

    A re-install will do nothing to change an issues that is hardware related. If you do check these things and see no issues, you may still consider reporting the hardware issue to the datacenter and see if they can test or replace the hardware for you.

    Cheers!

    Thanked by 2Mark_R jar
  • @TheLinuxBug said:
    Before re-installing I would highly suggest you use smartctl to check the status of your hard drive. If it is having a lot of failures this could explain the extremely slow IO speed. This article should give you the generics in doing this check: Click Here

    If you can confirm its not the drive starting to die, then I would make sure to review your system message log (dmesg) and confirm you see no other hardware giving errors, such as the SATA controller on the server.

    A re-install will do nothing to change an issues that is hardware related. If you do check these things and see no issues, you may still consider reporting the hardware issue to the datacenter and see if they can test or replace the hardware for you.

    Cheers!

    I read your comment after I reinstalled the server.. however, this server seems fine, work fast.. I will monitor for 2 or 3 days and will try your suggestion if the problem happen again.. thanks !

  • jarjar Patron Provider, Top Host, Veteran

    Mount with noatime, set noop scheduler. You may even have a drive going bad.

  • well, until now, my server working great..

  • mtr/traceroute your server, I have a server with online.net insane packetloss @ proxad (spelling of name, at work so don't recall exactly) network.. like 80% or more, even SSH was impossible

  • if w/a is high , i have always seen that there is some problem with with I/O of the machine , disk problems.

  • krs360 said: mtr/traceroute your server, I have a server with online.net insane packetloss @ proxad (spelling of name, at work so don't recall exactly) network.. like 80% or more, even SSH was impossible

    Might help if you actually look at what his problem is. packet loss probably won't cause a load in the 70's and nearly 80 iowait....

Sign In or Register to comment.