Edited Jan 23 to include code delimiters.
Sorry for the long post.
From about one year ago, after a major upgrade to our networking infrastructure, I've been documenting cleaning and tweaking the mess left by the company which did the job. I've reached a level where almost everything is well known, organized and under control. Now I'm studying and gathering info to go deeper into the switch core, which is (are) two OS6900-T40 operating in virtual chassis mode.
I'm running on 8.4.1.229.R02 GA
Problem 1:
My major concern are that PMD files are being generated, meaning that something bad is going on, which so far I don't know what is. I scheduled a system shutdown for next tuesday night when I intend to reboot the thing and see what happens after that. If PMDs continue to show up I'll open a support case with A-L.
I noticed that master and slave have significant difference on uptimes, the shorter (chassis 1) kind of matching a general power shutdown we did for maintanance on our 69 kV substation. The UPSs in theory held well during the outage, but for some reason chassis 1 shut down or rebooted.
Could it be that this is the cause for the PMDs? I erase the old ones to keep flash space below the threshold, but new ones continue to show up. The switch is not rebooting, at least not continuously:
Chassis 2: 311 days 20 hours 42 minutes and 16 seconds
Chassis 1: 98 days 7 hours 42 minutes and 46 seconds
Chassis 2 became master after chassis 1 rebooted.
Any experiences you wanna share about this? Should I try to reboot, or am I at risk of the switch not coming up again? That would be baaaad.
Problem 2:
I have two static routes in the configuration:
Code: Select all
! IP Route Manager:
ip static-route 0.0.0.0/0 gateway 10.1.119.251 metric 1
ip static-route 10.1.0.0/18 gateway 10.1.119.251 metric 1
Later, in scenario 2, internet access was shifted to flow through our central administration, thus sharing the route through 10.1.119.251.
Could it be that the person who configured these routes forgot to remove the first static route (10.1.0.0/18) after scenarios changed from 1 to 2? It's very likely given his profile. I'm not a routing expert, but it seems to me that 0.0.0.0/0 includes 10.1.0.0/18. Does it make sense to have both?
Thank you so much for any reply.
Tales Maschio
