slowing down everything soyoustart

Dilstar

Hi,

I am facing this issue since yesterday.
iotop result.

975 be/3 root 0.00 B/s 15.40 K/s 0.00 % 99.99 % auditd
18672 be/4 root 0.00 B/s 3.85 K/s 0.00 % 99.99 % sshd: root [priv]
13743 be/4 root 0.00 B/s 0.00 B/s 0.00 % 99.99 % [kworker/u24:2]
18738 be/4 mysql 0.00 B/s 0.00 B/s 0.00 % 99.99 % mysqld --defaults

how to resolve this issue and what it could be.

thanks

jar

I don't see a problem from this output. What were the conditions that caused you to run iotop and grab this output?

Dilstar

CPU usage too high,

wa is also too high. normally CPU goes above 50 mark easily. and remain there.

i am unable to do anything due to this issue.
i am using virtualizor with 1 container in it.

thanks

Advin

Check your CPU clock speed/frequency and CPU temperatures. Could be a hardware issue.

Void

SoYouStartSlowly
SlowYouStart

dahartigan

YouSoSlow

Dilstar

The drive with serial number: K5G is defective and should be replaced.

I have 4x 2To HDD SATA Soft RAID

if i tell them to replace the drive, so would be data be safe? and what procedure i have to do as its soft raid, after changing the drive.

or i just tell them to replace to make it fix ?

Thanks

momkin

@Dilstar said:
or i just tell them to replace to make it fix ?

No , if they replace the disk the server won't start , you will have to rebuild the raid manually from rescue system.

Ovh won't help with that !

Dilstar

after their intervention server node is fine, but i did not start the container/vps yet. if it works fine or not, WA wait time is also fine. i am thinking to start container and try to backup.

and here i dont know how to rebuild raid. as i did not encounter such issue before. i have no knowledge regarding rebuild the raid.

thanks

RapToN

I would recommend to Check your RAID status:
cat /proc/mdstat

Google for mdadm if you need to add the New disk to the RAID.

Dilstar

here is the result currently

[root@bigserver ~]# cat /proc/mdstat
Personalities : [raid1]
md4 : active (auto-read-only) raid1 sda4[0] sdb4[1] sdc4[2]
1837642688 blocks [4/3] [UUU_]
bitmap: 0/14 pages [0KB], 65536KB chunk

md2 : active raid1 sdd2[3] sdc2[2] sdb2[1] sda2[0]
104856512 blocks [4/4] [UUUU]

unused devices:

adns

@Dilstar said:
after their intervention server node is fine, but i did not start the container/vps yet. if it works fine or not, WA wait time is also fine. i am thinking to start container and try to backup.

and here i dont know how to rebuild raid. as i did not encounter such issue before. i have no knowledge regarding rebuild the raid.

thanks

Here is a good description by OVH https://docs.ovh.com/ie/en/dedicated/raid-soft/
The only that I make different that I use gdisk and generate new UUIDs on the new disk
sgdisk /dev/sda -R /dev/sdb # copy sda partition table to sdb
sgdisk -G /dev/sdb # generate new UUIDs

And double check the command before issue (E.G. disk identifiers, RAID array identifiers and more) because it is easy to damage data. Remember that you might have more than one raid arrays (for example if you use disk encryption with SSH unlock and you place the /boot on a different array).

Good luck:)

Dilstar

currently wa looks fine, did they remove that device which were causing issue or still there. as server looks ok curently, should i start my vps container and try to backup, as i dont have backup. or tell them to replace ?

thanks

Falzo

Seems like it is all in raid 1 aka mirroring. Means you could probably lose two more.disks and still be fine.

Doing backups right now is the roght thing to do..

Does not look like they replaced the disk at all. More likely only removed it from the raid array md4

Dilstar

hi,

i have bootup the container, wa is still high. but i am trying best to take backup.
looks like that drive is still exist there, which is causing issue to wait time.

thanks

Falzo

Check what smartmontols give there. If you rebooted the md4 raid might have been reassembled with the 4th disk and is running a rebuild.

Get a (paid) sys admin to analyze an fix it for you.