SG300 1.2.5.70 Spontaneous Crash / Restart

[cisco.mark]

2147483647     2012-Jun-27 20:23:40    Emergency     %OS-F-BUFFER: OSBUFG_buf_alloc: Buffer pool magic is invalid ***** FATAL ERROR *****  Reporting Task: GOAH. Software Version: 1.2.5.70 (date  11-Jun-2012 t ime  17:35:31) <snip> ***** END OF FATAL ERROR *****   

The switch was running a relatively basic configuration and not under any appreciable load when the crash occurred. It had been running 1.2.5.70 for quite some time and had been restarted several times since the firmware upgrade. I have not been able to reproduce the issue on demand.

[Ben M Johnson]

SG300-28p 1.2.5.70 - ours started rebooting every 24 hours give or take 15 minutes. 

We reverted to 1.1.2.

Waiting this one out.

[Tom Watts]

Hello cisco.mark and Ben Johnson,

Currently I do not see any bugs in regards to a spontaneous reboot/crash reported. I have been testing the SG300-28p switch on the 1.2.5.70 and haven't ran in to spontaneous drops or lock up yet. On my lab switch, a factory reset was performed after the upgrade. This is a recommended best practice.

I have 1 service request the customer has 9 switches, 3 of them have exhibited a similar behavior with rebooting. This customer upgraded their firmware on Saturday 6/30 and the 3 switches rebooted a total of 5 times.

On this scenario there was no fatal errors, port security, spanning tree, POE, EEE issues. At this time, we backed up the switch config, factory reset the switch, reloaded the config and we are now monitoring to see if there is a stability issue.

The 1.2.5.70 was a pretty significant coding implementation and I think it is possible there might be some lingering effect from the new software over the old.

I will update this post when my current open case shows any different result/symptoms over the next few days.

[Tom Watts]

Hi guys, please continue to use the 1.1.2.0 firmware. The business unit has recalled the 1.2.5.70 firmware.

Please reference the following;

https://supportforums.cisco.com/docs/DOC-25157