SMART error (CurrentPendingSector) detected on host

[jtvdw]

Hi,

We have a cluster of 5 nodes and from the 1 node and for the last 2/3 days we're been getting this error: (Running PVE5.2 on all of the nodes)

=====
This message was generated by the smartd daemon running on:

host name: server
DNS domain: domain.tld

The following warning/error was logged by the smartd daemon:

Device: /dev/sda [SAT], 1 Currently unreadable (pending) sectors

Device info:
WDC WD10EFRX-68FYTN0, S/N:WD-WCC4J7VJR4LZ, WWN:5-0014ee-20fb02337, FW:82.00A82, 1.00 TB

For details see host's SYSLOG.

You can also use the smartctl utility for further investigation.
The original message about this issue was sent at Sat Sep 14 16:56:39 2019 SAST
Another message will be sent in 24 hours if the problem persists.
=====

I've tried a couple of this like:

1. Forcing a fsck with touch /fourcefsck
2. Forcing a fsck in the grub but wasn't successful.
3. Manually forcing a fsck with fsck -fy /dev/sda1,2,3 but also not successfully.

Any help will be appreciated.

Thanks.

 

[ness1602]

I would change the disk ASAP.
I this is your production,you don't repair disk,you replace them fast as possible.

 

[jtvdw]

Thanks for the update, yes it's a production node

 

[jtvdw]

Hi,

HDD replaced and issues sorted.

Thanks.

 

[Jarvar]

I just received a similar error. However, this is a relatively new system and new disk.

Any suggestions? The disk in question is only a few months old and reports 87% life remaining.

I got this email last night
The following warning/error was logged by the smartd daemon:

Device: /dev/sdc [SAT], 1 Currently unreadable (pending) sectors
I logged into the server and ran smartctl -a /dev/sdc on the disk and this is the output. I don't see any errors or pending sectors...

 

[ness1602]

This is what type of ssd?

 

[vikozo]

@Jarvar which command do you use to display this information?
have a nice day
vinc

 

[Jarvar]

@Jarvar which command do you use to display this information?
have a nice day
vinc

I used smartctl -a /dev/sdc

 

[Jarvar]

This is what type of ssd?

I am using a SATA ssd by Crucial. The MX500 in 1 TB. It is in raid zfs 1 mirror configuration with a 1 TB Adata Xpg sx8200 Pro.

 

[ness1602]

Crucial is home-type ssd,not suitable for production(atleast i wouldn't use it). So realllocation and similar errors should occur even every week( Proxmox writes alot).

 

[Jarvar]

Crucial is home-type ssd,not suitable for production(atleast i wouldn't use it). So realllocation and similar errors should occur even every week( Proxmox writes alot).

Thank you. So this is normal for home-type ssds? I have been using this since October 2019. I have read on here several other people have used the MX500s..

I am setting up a new system using enterprise SSDs INTEL D3-S4610. 2x 240GB for the OS and 2x960GB for the VM Storage. I've built it up as new and setting it up, just trying to iron out a few kinks like getting updates, and wondering if I should copy the config files from the server it would be replacing.

Would you recommend provisioning extra space on the OS drives? or just use the automatic install with ZFS Raid 1 mirror? I've just gone through with the automatic process so far.
Thanks.

 

[ness1602]

I use sas for OS systems, and SSDs for vms(Seagate Ironwolfs, DC4610 ,and some older micron i guess). As for the recommendation, i'm not quite sure, i guess you could leave it working like that.

 

[Jarvar]

This error has disappeared, does that mean for the time being things are okay? Or is the system constantly ignoring any pending errors or sectors?

 

[ness1602]

If even one error occurs you should replace it. Thats my opinion.

 

[Jarvar]

@ness1602
Well, the thing is. I am setting up a new server so it takes time. I'm just trying to figure out how much time I have roughly. In the event the new server is setup, But I am just configuring settings. I'm also trying to figure out if I could setup a cluster, add both nodes and then "migrate" all the information from the old server over to the new one. I am trying to figure out the sense of urgency.
Also if there is one error, should I remove that drive? or the whole server? well probably just the SDD. I can put in an Intel D3-S4610 eventually and see how that runs...

On my home system, not even running proxmox, I have 1 of these which I'm not sure of

199

Ultra-DMA CRC Error Count

1

Errors

 

[ness1602]

Hm, ithink crc errot count usually points to bad sata cables,but not sure.