Last weekend when my parity check was running a SMB network share died as my old Unraid server (192.168.1.99) went down.  At the same time my current Unraid server, named Portrush, started having disk read errors on my parity drive (disk0).  These errors only lasted for a second at Dec 21 09:29:42.

 

Since then I don't see any new read errors, and I successfully went through a parity check.

 

Is this indicative of a problem on the parity drive or are these read errors possibly related to the other server going down.  This other server is set up as a SMB share on my main server.

 

On my daily notice report I still see the following:  "Parity - ST8000DM004-2U9188_ZR11R018 (sdb) - active 32 C (disk has read errors) [NOK]".  How do I reset this error?  Or should I?

 

See my logs below (logs truncated as there were 169 of these read errors).

 

Dec 21 09:29:39 Portrush kernel: CIFS: VFS: \\192.168.1.99\media BAD_NETWORK_NAME: \\192.168.1.99\media
Dec 21 09:29:39 Portrush kernel: CIFS: VFS: reconnect tcon failed rc = -2
Dec 21 09:29:41 Portrush kernel: CIFS: VFS: \\192.168.1.99\media BAD_NETWORK_NAME: \\192.168.1.99\media
Dec 21 09:29:41 Portrush kernel: CIFS: VFS: reconnect tcon failed rc = -2
Dec 21 09:29:42 Portrush kernel: ata1.00: exception Emask 0x0 SAct 0x3f80 SErr 0x0 action 0x0
Dec 21 09:29:42 Portrush kernel: ata1.00: irq_stat 0x40000008
Dec 21 09:29:42 Portrush kernel: ata1.00: failed command: READ FPDMA QUEUED
Dec 21 09:29:42 Portrush kernel: ata1.00: cmd 60/40:38:d0:98:f8/05:00:0b:02:00/40 tag 7 ncq dma 688128 in
Dec 21 09:29:42 Portrush kernel:         res 43/40:40:e8:98:f8/00:05:0b:02:00/00 Emask 0x409 (media error) <F>
Dec 21 09:29:42 Portrush kernel: ata1.00: status: { DRDY SENSE ERR }
Dec 21 09:29:42 Portrush kernel: ata1.00: error: { UNC }
Dec 21 09:29:42 Portrush kernel: ata1.00: configured for UDMA/133
Dec 21 09:29:42 Portrush kernel: sd 1:0:0:0: [sdb] tag#7 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Dec 21 09:29:42 Portrush kernel: sd 1:0:0:0: [sdb] tag#7 Sense Key : 0x3 [current] 
Dec 21 09:29:42 Portrush kernel: sd 1:0:0:0: [sdb] tag#7 ASC=0x11 ASCQ=0x4 
Dec 21 09:29:42 Portrush kernel: sd 1:0:0:0: [sdb] tag#7 CDB: opcode=0x88 88 00 00 00 00 02 0b f8 98 d0 00 00 05 40 00 00
Dec 21 09:29:42 Portrush kernel: I/O error, dev sdb, sector 8790776016 op 0x0:(READ) flags 0x0 phys_seg 168 prio class 2
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775952
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775960
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775968
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775976
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775984
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790775992
Dec 21 09:29:42 Portrush kernel: md: disk0 read error, sector=8790776000

 

It's logged as a disk problem, but these can be intermittent sometimes, and if you did a parity check after that, the disk is OK for now, keep monitoring.

Thanks Jorge, I will keep a close eye on it.

@JorgeB When I received my daily dis report it is still showing errors.  How do I reset the read error?

 

 

Event: Unraid Status
Subject: Notice [PORTRUSH] - array health report [FAIL]
Description: Array has 5 disks (including parity & pools)
Importance: warning

Parity - ST8000DM004-2U9188_ZR11R018 (sdb) - active 31 C (disk has read errors) [NOK]
Disk 1 - ST4000DM004-2CV104_ZTT2QHX8 (sdc) - active 34 C [OK]
Disk 3 - ST8000DM004-2U9188_ZR11SQ9R (sdd) - active 31 C [OK]
Cache_nvme - WDC_WDS500G2B0C_21305D807219 (nvme0n1) - active 38 C [OK]
Cache_nvme 2 - CT500P3SSD8_2310E6B819BE (nvme1n1) - active 35 C [OK]

Parity is valid
Last checked on Sun 29 Dec 2024 08:15:15 AM EST (yesterday), finding 0 errors.
Duration: 7 hours, 15 minutes, 9 seconds. Average speed: 306.5 MB/s

52 minutes ago, wayner said:

When I received my daily dis report it is still showing errors.  How do I reset the read error?

Reboot, or re-start the array, not sure if the latter is enough.