Card slots not coming online
a_riot42
Posts: 1,624
in AMX Hardware
At two different sites I have 4100s that once every few reboots won't allow the card slots back online. A reboot or two usually fixes this but its a major annoyance. Has anyone seen this behavior? I reported it to AMX but they want more debug which I can't give them as there are clients living in the locations where this is happening.
Thanks,
Paul
Thanks,
Paul
0
Comments
With the 3100, it seemed like the only way to get the ICSNet back was to pull to power to the processor for a few seconds, altho once I had it work with a soft reboot. On the 4100, I have had it work with a reboot.
Jeff
P.S.
I have reported it to tech support, but it has been so difficult to reliably reproduce that I have not expected an answer to be found quickly.
I have only seen it on the 4100s and a reboot usually solved it. I have watched the startup sequence and noticed no difference between a boot when they come online or not. Its a major bug though, since having no cards work disables the system until its rebooted.
Paul
Yup - we experience the same thing quite often with 4100s, but also with 3100s as well (Axlink bus strip is usually connected, and those won't come online.)
I've learned to monitor the devices in the card frames and after 10 or 15 seconds (in DEFINE_START) check to make sure they're all online, if not - I send a reboot. If it reboots 5 times without them coming online, I put up a popup on the panels to have them call use immediately to remedy the situation. The latter hasn't happened yet, but just a bit of a CYA.
Another thing that helps is telnetting into the master and setting the Lontalk threshold to the max. For the 4100s it defaults to something really low like 50.
I wasn't involved from the start, but perhaps my situation started off like what you are seeing, Paul?
It was pretty intermittent, I haven't seen it in the past month or so.
A power cycle or another reboot always fixed it.
I wonder if AMX has done anything to address this.
I've been fighting this since July of 2007 (my first time seeing it was then.) I know I had reported it then and spoke to tech support at length about it. I've come to just accept it and check to make sure they're online before proceeding.
That seems to the reasonable approach, but even needing something like that is rather lame. What's next, making sure the built in ports are online too?
Paul
Agreed but . . . oh well.
I'm still waiting for cruise control to actually take me to my destination.
Make that seven.
The only issue I've ever had with the many 4100s I've put in was more of an overall ICSNet dropping offline. I was never able to isolate it to the card slot within the 4100 itself. So, I'll make it 7.5 people who've had trouble.
e
Just my hunch.
I've had the same problem, I ended up rolling back the firmware on a couple projects to make it work every time. Some one in tech supports does know about it, because they gave me the older firmware.
problem ++;
http://www.amxforums.com/showthread.php?p=36121#post36121
I'll implement James' track/reboot fix, but curious if AMX ever found an issue.All the latest firmware is installed.
Kevin D.
Jeff
I reboot the master 3 times if still no workie. after 3 strikes, I get an email and page to my phone that trouble awaits. This is in two systems with 4100s.
Jeff
I've had the problem in the past but haven't noticed it lately either. Usually they would come up within a few reboots.
--John
Kevin D.
I have seen this as well, where all 4 cards were not online. I have also experienced a situation where none of the Duet devices ever came online. The reboot fixes the problem, but it makes for a real flakey system when the homeowner only sees that his expensive system isn't working!
This will help in SOME situations.
Force the master to wait before loading data_event online data
TELNET: set device holdoff on
query: get device holdoff
I have this problem with a 4100 as well, both the cards not online and the Duet virtuals not online. So basically, 3/4 of there system doesn't work. The reboot works to refresh, but it should just F-ing work the first time. This is like starting your car and none of the gauges work, and having to re-start your car to get the speedo to report your speed. I am getting tired of the AMX BS, with Vapor-Firmware that will someday fix the problem, constantly hearing from TS to update firmware, only to have other issues/problems introduced with the new firmware.
I can't be the only one that feels like AMX quality control is out of control?!?
This has side effects as well though doesn't it? Doesn't it disable all messages during define_start?
Paul
Considering the breadth of their product line, problems like this are few and far between. 95% of the AMX products I have dealt with over the years were rock-solid ... of course, that remaining 5% stands out because they brings the whole system crashing down. The car analogy is apt there too ... the owner doesn't care if it's a minor issue or a major one that prevents his car from starting, and doesn't appreciate one bit how hard it might be to track down the problem. He just wants to use his car. I have a friend who ranted and raved about his Saturn (before GM ditched them) when the engine died on him ... how he would never buy another "piece of junk" Saturn again, etc., etc., what an unreliable car it was. Yet, I had one myself that was over 10 years old, was never in the shop for anything more than an oil change, and even still looked new. No matter how good your overall product line might be, if you have a few glaring issues, it those that people notice and remember.