Ticket #794 (closed task: fixed)

Opened 10 years ago

Last modified 10 years ago

network problems with Sieg testbed systems

Reported by: jsamuel Owned by: monzum
Priority: critical Milestone:
Component: sysadmin Version: 0.1o
Severity: Medium Keywords:
Cc: justinc Blocking:
Blocked By:


The computers in Sieg (I believe it's the ones in Sieg) seem to have started having network problems on Friday evening. The continuous build fails on these systems starting then and the entire weekend I've had very slow and flaky ssh to testbed-opensuse and testbed-mac while trying to find out more.

I notice that sometimes I get partial packet loss on a ping to testbed-opensuse (other times I get full packet loss):

$ ping -c 4 testbed-opensuse
PING testbed-opensuse.cs.washington.edu ( 56(84) bytes of data.
64 bytes from testbed-opensuse.cs.washington.edu ( icmp_seq=2 ttl=63 time=0.366 ms
64 bytes from testbed-opensuse.cs.washington.edu ( icmp_seq=3 ttl=63 time=0.340 ms
64 bytes from testbed-opensuse.cs.washington.edu ( icmp_seq=4 ttl=63 time=0.346 ms

--- testbed-opensuse.cs.washington.edu ping statistics ---
4 packets transmitted, 3 received, 25% packet loss, time 3011ms
rtt min/avg/max/mdev = 0.340/0.350/0.366/0.024 ms

That's also quite a high ping time as I ran this from on campus.

The monitoring system has also been reporting to our irc channel failures all weekend:

2009-12-14T07:25:23  <Seattle_TestBot> WARNING: Seattle machines are down!
Seattle developers please check on the machines.
2009-12-14T07:25:24  <Seattle_TestBot> Displaying ping result for ping failures:
2009-12-14T07:25:27  <Seattle_TestBot> Pinging testbed-freebsd.cs.washington.edu: packets received 0%
2009-12-14T07:25:27  <Seattle_TestBot> Pinging testbed-xp2.cs.washington.edu: packets received 0%
2009-12-14T07:25:27  <Seattle_TestBot> Pinging testbed-vista1.cs.washington.edu: packets received 0%
2009-12-14T07:25:28  <Seattle_TestBot> Pinging testbed-opensuse.cs.washington.edu: packets received 0%

We should figure out as soon as we can if this is fixable by us and, if not, open a ticket to lab.

Change History

Changed 10 years ago by monzum

Yes I noticed this problem on Friday afternoon. The activity lights keep going on and off on the router that all the machines are connected through, which is causing the network connectivity problem. I believe its a problem with the router and not the internet connection to the room. I've already contacted cs.support about this.

Changed 10 years ago by monzum

  • status changed from new to closed
  • resolution set to fixed

CS support has replaced the router in the Sieg lab. There hasn't been anymore fail notification since then. If this problem resurfaces, then the ticket could be re-opened.

Note: See TracTickets for help on using tickets.