Jay Ashworth
Well-Known Member
Just cleared a network problem, and I thought everyone might enjoy a nice saga...
We've been having a problem, the last month or so, where our Production LAN fell off the Internet -- I have ScreenConnect set up to get me to the two Mac Mini's, one in each booth, in case I have to support something, and each of them would disappear somewhere in the 9am hour weekdays, and come back on somewhere between 2 and 5pm.
If you think that sounds human-driven, you have much to go on, but *I* didn't get it until today, while the campus IT guy was pedaling over here with all his gear...
It turned out that the root cause was that some time During The Emergency, someone had unplugged the control Ethernet cable from the relevant jack on the back of the A&H Qu-16 desk...
... and plugged it back into the *D-Snake* jack. An Ethercon style jack, not a standard box jack. Me and my supervisor are the only 2 people who could reasonably be expected to unplug that wire, and we're both pretty sure we didn't, so it was clearly Bil Keane's kid, Ida Know.
That rack was only powered on for those 4 weeks ... weekdays, approximately the times that we saw the network go down. (We had a dance company renting it for intensives over the summer.)
No way to tell how long it was patched wrong. It only caused a problem while it was powered up.
What I'd love to know, though, is why patching a D-snake jack into the production LAN took down the whole LAN, including the upstream router (an ASUS RT-AC86U). Took it down so bad that SACN commands from our Ion to our movers executed... but only seconds (or minutes) later. Anybody sufficiently familiar with the D-Snake to speculate on why that connection might have caused such a foofaraw?
We've been having a problem, the last month or so, where our Production LAN fell off the Internet -- I have ScreenConnect set up to get me to the two Mac Mini's, one in each booth, in case I have to support something, and each of them would disappear somewhere in the 9am hour weekdays, and come back on somewhere between 2 and 5pm.
If you think that sounds human-driven, you have much to go on, but *I* didn't get it until today, while the campus IT guy was pedaling over here with all his gear...
It turned out that the root cause was that some time During The Emergency, someone had unplugged the control Ethernet cable from the relevant jack on the back of the A&H Qu-16 desk...
... and plugged it back into the *D-Snake* jack. An Ethercon style jack, not a standard box jack. Me and my supervisor are the only 2 people who could reasonably be expected to unplug that wire, and we're both pretty sure we didn't, so it was clearly Bil Keane's kid, Ida Know.
That rack was only powered on for those 4 weeks ... weekdays, approximately the times that we saw the network go down. (We had a dance company renting it for intensives over the summer.)
No way to tell how long it was patched wrong. It only caused a problem while it was powered up.
What I'd love to know, though, is why patching a D-snake jack into the production LAN took down the whole LAN, including the upstream router (an ASUS RT-AC86U). Took it down so bad that SACN commands from our Ion to our movers executed... but only seconds (or minutes) later. Anybody sufficiently familiar with the D-Snake to speculate on why that connection might have caused such a foofaraw?