Unschedueled/Unexpected Downtime

Post by **porcupine** » Wed Aug 07, 2002 12:49 am

Well folks,

It seems our first unschedueled downtime has come and gone. The network was temporarily unavailable 40-50 minutes.

The local Cogent router failed (reason unknown) and the people at Ultranet were called down to assist Cogent as Cogent's remote hands to speed up the process instead of having a Cogent technician having to get onsite to do such labour.

If you've got any questions, let me know. I will be investigating this mater further to some extent to see if i can ferrit out the reason that the cogent hardware had failed as they utilize cisco 15434 routers, which are nothing short of top of the line.

Regards,

Post by **porcupine** » Wed Aug 07, 2002 10:25 pm

Ok,

Well the good news is, we've found out what the issue is. The master Cogent Cisco Switch, and has failed, and is being replaced tomorrow sometime as they are rushing out and currently configuring a new one to minimize downtime.

There will be obviously a small amount of downtime tomorrow, semi-schedueled as they swap the old switch out and put the new one in. There aren't many connections or configurations on it so it shouldn't take less then 30 minutes we hope.

solatis · Post by **solatis** » Thu Aug 08, 2002 9:05 am

Any idea when these problems will definately be fixed?

After tomorrow?

Post by **porcupine** » Thu Aug 08, 2002 1:30 pm

Should be if the network engineers analysis of the problem was correct. Cogent has not failed on this location for 8 months prior to this having minimal unschedueled downtime (under 1 hour total for the previous 8 months), and then suddenly there are 3 spikes of downtime in the past week, so it's obviously a new problem, and hopefully they've analysed the source correctly, and replacing this switch will do the trick.

I'm sure if they're not correct, the problem will re-occur several more times and they will have to re-analyse the situation.

Post by **porcupine** » Fri Aug 09, 2002 11:55 pm

i believe that was the switch replacement, that 10 minutes of downtime, but i'm still waiting for confirmation as it was the cogent people doing this maintenance, not our own.

I certainly hope thats what it was though, because thats around the right amont of time (if you've ever started up a cisco switch, it takes around 5 minutes for the thing to boot and check every port as it starts up).