|
Joined: Jan 2005
Posts: 264 Likes: 1
Member
|
Member
Joined: Jan 2005
Posts: 264 Likes: 1 |
Have you guys run into DX-80s where the programming breaks? The tech that taught me much of my phone skills always blamed corruption. I don't feel like I've ever had a DB that was broken. Usually just cold-start and re-upload and it clears software glitches, or proves that it's hardware.
About a month ago I had a DX-80 have board #2 DPM8 go offline. Thought the card was dead, but swapping it didn't help. Ended up cold-starting, and restoring programming and all was well. I even put back the original DPM8. So, get a call this morning that the one half of the business can't access CO lines. Had them verify the modem was online, but still not go after rebooting it. Have her check a station at it says CO GROUP UNAVAILABLE when she hits 9 for a line, and dead silence when she attempts to hit a specific line. I assume that this is the same issue that we had almost a month ago - - May 29th and today is June 25th. I'm thinking a cold-start will solve it - - unless it really is the cable modem that shows as online. The call last month was the first peep I've heard from this customer probably in 18 months. If I can remember there was a few times we had them reboot the cabinet for a stuck VM or something in the last few years.
So here's my question: When do you say it's hardware failing (CPM & Backplane) vs software? I'm leaning toward swapping the hardware, but I think a cold-start and sending the last DB will probably bring everything back. I feel like a DX-80 doesn't like power fluctuations / brown out / and other summer storm related issues. We've had tons of rain & t-storms in the past few months. Starting to feel like monsoon season around here.
Last edited by tylan; 06/25/18 10:18 AM.
|
|
|
Visit Atcom to get started with your new business VoIP phone system ASAP
Turn up is quick, painless, and can often be done same day.
Let us show you how to do VoIP right, resulting in crystal clear call quality and easy-to-use features that make everyone happy!
Proudly serving Canada from coast to coast.
|
|
|
Joined: Dec 2007
Posts: 6,807 Likes: 18
Retired Admin
|
Retired Admin
Joined: Dec 2007
Posts: 6,807 Likes: 18 |
The tech that taught me much of my phone skills always blamed corruption....Usually just cold-start and re-upload and it clears software glitches, or proves that it's hardware.
Ended up cold-starting, and restoring programming and all was well. It's not clear regarding your method of cold-start and re-upload. If there is indeed a corruption in the database and if you download the database, cold-start then upload the data base, all this does is reintroduce the original problem...uploading a corrupted database. Proper method:Download the database. Hand copy the database, screen by screen to a different machine (laptop). Cold-start the KSU as per mfg instructions. Upload the clean version of the database. It could be that you are following this methodology but your post isn't clear on that point.
|
|
|
|
Joined: Jan 2005
Posts: 264 Likes: 1
Member
|
Member
Joined: Jan 2005
Posts: 264 Likes: 1 |
I wasn't really clear on that. When I say restoring after a cold start, that is typically a backup that I had on my laptop. Usually from the last visit I was there. I have already downloaded the database, cold started it, and uploaded the same one and seen that work just fine. I have also done the screen by screen rebuild of the database.
Also your reply indicates that you do believe that actual corruption can take place. That's the part I never really been sure about. if it was an old technician's tale or if it was a true issue. It usually seems to be hardware.
Last edited by tylan; 06/25/18 11:44 AM.
|
|
|
|
Joined: Mar 2018
Posts: 345 Likes: 1
Member
|
Member
Joined: Mar 2018
Posts: 345 Likes: 1 |
I have had issues where the analogue card was the cause of problems. Something will happen to the card and it will pop, then you have random ghosts in the system causing issues. A sign of this is the cabinet will be very hot, and/or the plastic around the heatsink on the analogue station card will be melted to discoloured.
I have also had problems like you are talking about. In those cases I will swap out the backplane and the cpm and then send their old ones in for repair. (although those are usually full up dx120s)
Yes the dx80 does not like power fluctuations. Very few of our customers like to pay for a battery backup, until we have been out a couple of times to fix the kinds of issues that you are talking about.
The Cloud is just someone else's computer!
|
|
|
|
Joined: Jan 2005
Posts: 264 Likes: 1
Member
|
Member
Joined: Jan 2005
Posts: 264 Likes: 1 |
Customer said they had power issues recently.
It's on a battery backup which looks fairly new.
Correction, it said "CO GROUP BUSY" when you hit 9. Not co group unavailable. Rebooting the cabinet took care of the issue.
|
|
|
|
Joined: Dec 2007
Posts: 6,807 Likes: 18
Retired Admin
|
Retired Admin
Joined: Dec 2007
Posts: 6,807 Likes: 18 |
I wasn't really clear on that. When I say restoring after a cold start, that is typically a backup that I had on my laptop. Usually from the last visit I was there. I have already downloaded the database, cold started it, and uploaded the same one and seen that work just fine. I have also done the screen by screen rebuild of the database.
Also your reply indicates that you do believe that actual corruption can take place. That's the part I never really been sure about. if it was an old technician's tale or if it was a true issue. It usually seems to be hardware. I have indeed seen database corruptions. On one particular system it would first show up in the system speed dial number. Example, instead of: No. Dir# ABBR. No Name TRK/RTE COS Override001 600 8005551212 Billy Bob Null N You would see something like this: N@. Dir# ~`#R. No Name TRK/RTE COS Override001 600 ®00555›212 Bil^5 Bob N±à °
|
|
|
|
Joined: Jan 2005
Posts: 264 Likes: 1
Member
|
Member
Joined: Jan 2005
Posts: 264 Likes: 1 |
I think I've seen stuff like that before. I think it was back on the DOS software. Not sure if I've seen it since the software was the windows based.
So, looking at this system it's plugged into the battery backup... But the surge only side! Not the battery side. LOL. Customer must have swapped the back ups and plugged it in wrong.
Also, when powering off the system a second time it lost the date and I think some programming. I bet the CPU battery, the one on the board is probably 10 years old. I bet that's part of what's going on... Power goes out for a few minutes and no battery to keep the time / programming.
|
|
|
|
Joined: Dec 2007
Posts: 6,807 Likes: 18
Retired Admin
|
Retired Admin
Joined: Dec 2007
Posts: 6,807 Likes: 18 |
Most UPS batteries last 3-5 years depending on various factors so the companies I have worked for test the Fire Alarm Systems and Security Systems every year and replaced every three years.
Why wouldn't you follow the same policy with a telephone system. UPS's are not from Ronco®, it's not "Set it and forget it".
Yeah, that was an example from the good ol' shell out to DOS days.
|
|
|
|
Joined: Jan 2005
Posts: 264 Likes: 1
Member
|
Member
Joined: Jan 2005
Posts: 264 Likes: 1 |
I tested the UPS for 5 minutes and it seemed OK. Plugging into the battery protected side is going to work much better! New UPS or not they gotta plug it into the battery protected outlets! 😂
I also swapped the CR2032 on the CPM card. I saw the system lose time I powered it off / on while testing. New CR2032 and it kept time when powered off for 60 seconds.
Seems like this could all be power related. Usually bad hardware is more easily reproduced.... I'm hoping there's no corruption, I sure don't see any signs of it.
|
|
|
|
Joined: Dec 2007
Posts: 6,807 Likes: 18
Retired Admin
|
Retired Admin
Joined: Dec 2007
Posts: 6,807 Likes: 18 |
|
|
|
Forums84
Topics94,384
Posts639,240
Members49,799
|
Most Online5,661 May 23rd, 2018
|
|
1 members (teleco),
91
guests, and
73
robots. |
Key:
Admin,
Global Mod,
Mod
|
|
|
|