You are not logged in.
Hello
I have 3 TB harddrive formatted with ext4, and mounted with "defaults,rw" and no pass or dump, which is all the time being read 2048 kb/s to about 4096 kb/s, it seem to most switch between those two rates, all the time while my filesystem which contains root is hardly ever reads more than 100 kb/s while the only software I run is default base with a minimal kde and chromium.
I run linux-ck from repo-ck with cfq.
Any ideas?
Last edited by Roberth (2013-06-10 17:36:02)
Use the Source, Luke!
Offline
So what part of the filesystem is it, and can you use something like iotop or lsof to determine what it is actually reading? I think you need to provide more information here... as saying simply that your HDD is readig a lot is not really anything to go on here.
Offline
"ext4lazyinit" is dissappearing and appearing with abotu 3 % all the time in iotop with the -o option, whatever that is?
Use the Source, Luke!
Offline
Use iotop and sort by disk reads:
sudo iotop -Pao
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
"jbd2/sdb2-8" is also at almost 1 % all the time, thats about it.
Use the Source, Luke!
Offline
"jbd2/sdb2-8" is also at almost 1 % all the time, thats about it.
By sorting by disk read or disk write you will see a cumulative listing to help you identify the culprit. For example, on my system:
Total DISK READ : 0.00 B/s | Total DISK WRITE : 0.00 B/s
Actual DISK READ: 0.00 B/s | Actual DISK WRITE: 0.00 B/s
PID PRIO USER DISK READ> DISK WRITE SWAPIN IO COMMAND
2610 be/4 facade 412.00 K 68.00 K 0.00 % 0.00 % libpepflashplayer.so --ppapi-flas~eccomp-sandbox --scroll-pixels=300
143 be/4 root 0.00 B 0.00 B 0.00 % 0.14 % [kworker/6:2]
164 be/4 root 0.00 B 156.00 K 0.00 % 0.00 % systemd-journald
1702 be/4 facade 0.00 B 12.00 K 0.00 % 0.00 % gnome-pty-helper
3313 be/4 facade 0.00 B 4.00 K 0.00 % 0.00 % zsh
894 be/4 root 0.00 B 0.00 B 0.00 % 0.00 % [flush-8:16]
399 be/3 root 0.00 B 100.00 K 0.00 % 0.01 % [jbd2/sda2-8]
951 be/4 facade 0.00 B 0.00 B 0.00 % 0.00 % xfdesktop
441 be/3 root 0.00 B 112.00 K 0.00 % 0.11 % [jbd2/sdb1-8]
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Yes, and I am reporting what I see which has significantly with IO activity. The rest is like around 0.01 % IO, the ext4lazyinit process doesn't read or write, it just produces IO activity, I guess it has something to do with the filesystem itself?
Use the Source, Luke!
Offline
No idea what ext4lazyinit is... take to google?
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Apparently, ext4 has a feature in which mkfs.ext4 doesn't actually finish, and the final non-essential parts are written after the first mount (or subsequent mounts if not enough time is allowed). It sounds as though Roberth is having this happen all the time, which doesn't seem normal.
Offline
Thank you for the good explanation, found the explanations I found on google to be, kind of unclear.
Well it happend just now when I created the filsystem, and it's 3 tb large, maybe that's the reason?
Use the Source, Luke!
Offline
That could very well be the reason. Leave it running and mounted and see if the chatter subsides after a bit. Maybe it is just checking over things to make sure all the necessary parts are written. A 3TB rotational disk is a lot of storage to cover and check over, so this sounds reasonable.
Offline
Yes, in the process when I resized the former filsystem on the partition, and created this one, and resized it to use the entire disk after I deleted the old one, took about 5 hours or so.
Use the Source, Luke!
Offline
Well when the night had gone by it was gone and IO was back to normal. But thanks, didn't even know iotop existed.
Use the Source, Luke!
Offline
Neato! Please mark the thread as [Solved].
Offline