Install of nodejs and npm very slow (very high IO delay)

[DST73]

Hi,

I have created a new CT but if I try to install npm it is VERY slow..... running now for almost 30 min
I have noticed that this isn't only on the new created but on others as well.

Looks like I have a very high IO delay


I have two drives.

the first drive I use for all my containers and backup
I have noticed that on my second drive the LVM is very full

How can I see, what is inside?

Could that be the reason why installing anything takes very long? Not only npm. For example I install webmin on some of my containers and that takes very long too.


Thank you!

 

[leesteken]

Your Samsung SSD uses QLC flash memory, which becomes slower than an old HDD if you write to them for a long time. Lots of threads (with disappointed people) on this forum about QLC drives.

 

[DST73]

Can you suggest any SSD?

 

[DST73]

What about that sdb3? How can I see why is inside? Thanks

 

[leesteken]

Can you suggest any SSD?

(second-hand) enterprise SSDs with PLP. (Did I mention already that there are lots of threads about QLC SSDs and advise about replacing them?)

What about that sdb3? How can I see why is inside? Thanks

I don't use it regularly but lvscan or some other LVM command can probably do that: https://pve.proxmox.com/pve-docs/pve-admin-guide.html#chapter_lvm

 

[DST73]

I found an advice to install iotop
Looks like my "Home Assistant" kvm is causing a high IO.
Could that be the problem ?

 

[DST73]

Hi,
a little update.
I switch to a new SSD and the IO delay is totally gone..
Is there anything I can do with my old disk?
Tried to reinitialize it and use it as backup but that caused as expected the high delay again.
Do I have to throw the disk in the trash?

Thanks!

 

[DST73]

Ok, tried my fist backup to then new SSD and the IO-deplay is back
Looks like it is only gone during the normal use.

It even cause the system to write the cpu graph

NFO: starting new backup job: vzdump 201 --storage ext_pve --compress zstd --node pve --remove 0 --mode snapshot --notes-template '{{guestname}}'
INFO: Starting Backup of VM 201 (lxc)
INFO: Backup started at 2023-11-02 10:44:23
INFO: status = running
INFO: CT Name: docker
INFO: including mount point rootfs ('/') in backup
INFO: mode failure - some volumes do not support snapshots
INFO: trying 'suspend' mode instead
INFO: backup mode: suspend
INFO: ionice priority: 7
INFO: CT Name: docker
INFO: including mount point rootfs ('/') in backup
INFO: starting first sync /proc/7970/root/ to /mnt/pve/ext_pve/dump/vzdump-lxc-201-2023_11_02-10_44_23.tmp
cannot remove directory for /mnt/pve/ext_pve/dump/vzdump-lxc-201-2023_11_02-10_44_23.tmp: Read-only file system at /usr/share/perl5/PVE/VZDump.pm line 1202.
ERROR: Backup of VM 201 failed - command 'rsync --stats -h -X -A --numeric-ids -aH --delete --no-whole-file --sparse --one-file-system --relative '--exclude=/tmp/?*' '--exclude=/var/tmp/?*' '--exclude=/var/run/?*.pid' /proc/7970/root//./ /mnt/pve/ext_pve/dump/vzdump-lxc-201-2023_11_02-10_44_23.tmp' failed: exit code 11
INFO: Failed at 2023-11-02 10:47:15
cp: cannot stat '/mnt/pve/ext_pve/dump/vzdump-lxc-201-2023_11_02-10_44_23.log': Input/output error
INFO: Backup job finished with errors
TASK ERROR: job errors

the backup ends with an error