Connectivity 1

[ITSmith]

Hello,

Not sure if this is the correct sub-forum to post to, but most of it is Cisco equipment in question.
Background:
- 15-17 users at this location at any given time.
- Flow is as follows:
T1 to Cisco Router
to Netgate
to Packeteer
to Gigabit switch
to Cisco 24-port hubs (3)
to Patch panels
to office ports.

Curious if anyone has witnessed similar symptoms:

Everyone went down before 9 A.M., then sporadic -
mostly down.
- Reboot router, Netgate, Packeteer, Gigabit, hubs
- From server in that room, we had connectivity to
Internet, ping corp srvrs, browse domain tree, etc.
- Lost srvr connection later
- Users still out
- Changed select cables one at a time
- Srvr had connection to outside again - thought it was the
last cable changed, but most users were still not able to
log onto domain (at remote corp office).
- After that last cable change, had connection from that
point on, to the outside with my 'Admin' pc, and another
IT machine.
- As I continued troubleshooting, over a period of 10s
of minutes to an hour or more, users were coming to me
saying that they were logging in, and now getting to the
outside.
- I saw the same from a few stations - first I couldn't get
out, then 5-10 minutes later I could.
- Etc.
- People had been trying repeatedly for some time, so it
sounds like something (DNS related?) Numbers being
flushed or otherwise refreshed over time from the
switch ? hubs ?
- Everyone is back up and getting to outside now,
but I'd like to get to the bottom of it.

Any thoughts would be greatly appreciated.

Thank You -
ITS

 

[BOKA]

I know of a similar situation where some weird problems were popping up from time to time and then disappearing without apparent reason. I think the issue turned out to be a BootP enabled on the 3COM managed ethernet switch, and it was grabbing IP address of a server which was Active Directory DC, DNS server, DHCP server, etc.
The exact scenario escapes me at the moment, but the solution was to go into switch's config and disable BootP.

If I recall additional deatails, I will post them here.

Good luck.

 

[ITSmith]

Thanks, Boka -

Will look into the BootP aspect of things.

As a side note:
The day after this originally happened (yesterday) most users logged in OK, but myself and one other staff member could not connect -
same symptoms as the day before.
We found his problem immediately -
He had a static address of xxx.xxx.0.1
When I reset it to obtain, he was OK.

Not sure how that happened, unless his home setup changed something - we've seen something similar before. A user used to travel and connect at one or more kiosks somewhere and almost every time he came back to the office from this place, he couldn't get to internet until we reset his properties. That was mainly when we had proxy, and his settings kept getting wiped out.

On mine, I set a static for myself, and was able to connect, but didn't have network drives since I hadn't actually logged in.
Later, I set it back to obtain, ran manual DNS flush, etc.,
That didn't do it, and I eventually ran the 'Repair' option for the connection, rebooted, got a valid address and was able to log in.

If I track it down, will post the findings.

Thank You Again -
ITS

 

[voltron1011]

You mentioned you have 3 24port Cisco Hubs? Can you verify this? Are they actually switches? If they are actual hubs then you might have some serious collision/bcast-storm issues on your segments.