Symptom
On N77-SUP2E supervisors, the logflash filesystem may suddenly go offline. The kernel messages in "show system internal kernel messages" will contain a trace similar to the following output, referring to USB device 2-1.3 and the block device sdc1:
[3786183.755587] usb 2-1.3: reset high-speed USB device number 5 using ehci_hcd
[3786198.834503] usb 2-1.3: device descriptor read/64, error -110
[3786214.013359] usb 2-1.3: device descriptor read/64, error -110
[ ... snip ... ]
[3786265.277428] EXT3-fs (sdc1): error: ext3_journal_start_sb: Detected aborted journal
[3786265.277501] EXT3-fs (sdc1): error: remounting filesystem read-only
The "show logging log" output will contain similar messages, followed by:
2018 May 23 14:39:02 N77 %USBHSD-2-UMOUNT: logflash: offline
Workaround
Perform a power cycle of the affected supervisor, either by physical re-seat, or by "out-of-service module" CLI command. A simple supervisor switchover or reload will not recover the logflash. If the logflash is not recovered even after a physical re-seat of the supervisor, it must be RMAed.
Further Problem Description
A firmware issue has been identified with the SMI SM3260 controller used on the logflash eUSB drive. Under circumstances, the firmware may access incorrect address on the drive, causing the drive to hang.
