Symptom
Dmesg reports I/O errors followed by remounting the filesystem as read-only:
Buffer I/O error on device sda4, logical block 0
EXT4-fs error (device sda4): ext4_journal_check_start:56: Detected aborted journal
EXT4-fs (sda4): Remounting filesystem read-only
EXT4-fs (sda4): previous I/O error to superblock detected
There is a possibility where switch can then proceed to crash due to bootflash going read-only
Conditions
A switch with an M1100 SSD running an affected version:
Affected M1100 Firmware versions:
- M0MU040 and below
Upgraded M1100 firmware version with fix:
- M0MU050
How to check if ACI switch has M1100 SSDs
moquery -c eqptFlash
---
leaf1# moquery -c eqptFlash
Total Objects shown: 1
# eqpt.Flash
acc : read-write
cap : 122099
childAction :
cimcVersion :
deltape : 1
descr : flash
dn : sys/ch/supslot-1/sup/flash
gbb : 0
id : 1
lba : 0
lifetime : 6
majorAlarm : no
mfgTm : 2023-12-20T10:39:59.805+09:00
minorAlarm : no
modTs : 2024-03-26T09:35:24.053+09:00
model : Micron_1100_MTFDDAV256TBN <<<<<<< Starting with "Micron_1100"
monPolDn : uni/fabric/monfab-default
operSt : ok
peCycles : 381
readErr : 0
rev : M0MU040 <<<<<<< Firmware version
rn : flash
ser : XXXXXXXXXXX
status :
tbw : 20.394547
type : flash
vendor : Micron
warning : no
wlc : 0
---
OR
smartctl -a /dev/sda (needs root permission, Cisco TAC need to be engaged)
Workaround
Reloading the switch should bring it back to full operation
Further Problem Description
Switches with M1100 SSDs can be run in either Standalone NXOS or ACI mode.
For Switches running NXOS/Standalone, refer to CSCwa86549
For Switches running ACI, refer to CSCwh70359
