I have a DX-120 installed about three months ago with the 7271C flash card that has crashed four times in the past three weeks.

When you dial the VM ports (157, 158, 159, and 160) or when dialing the hunt group 433, the VM doesn't answer...you only hear ring back.

The heart throb LED in the upper right corner of the board is still flashing and the flash card LED is still flickering. A tap on teh reset button restores service.

It happened two weeks ago, six days ago, three days ago and again today. After it crashed the third time, I replaced the VM unit with my spare one and swapped the flash cards to retain the prompts and messages, and set the corrected routing mailbox numbers.

The new unit just crashed today.

The DX-120 has a battery backup and they have a power outage for less than an hour at least every two weeks. I don;t know if there was an extended power outage the ran down the battery the first two times, but I do know that there were no power outages the last two times this week.

I don't think its a bad flash card, but what I'm going to do is reset the system and do an audio backup of the prompts (Audcity is a good audio editor) and stored messages (I don't want to back up the flash card per se as it might be corrupted). I have a adapter cable that plugs into the phones headset jack and my laptop.

If it stays up for at least two weeks, then it was a bad flash card. If it keeps dropping, I'll cold boot the system and reprogram it from a paper copy of the programming.

If THAT doesn't work, I'll replace the cabinet.

Any suggestions?

sounds like the VM is getting stuck in an endless loop rolling from port to port and tieing all the ports up. Next time it stops working, connect up with keylinkDX and see what the status of the VM is.
You may have a disconnect supervision problem on a line or a programming issue in a ucd group that sends the VM in to a loop when it is accessed.

That's what it sounds like to me too. I agree, I would be interested to see what the voicemail is actually DOING when it "locks up".

I went back this morning and the VM had fixed itself, so I didn't have a chance to look at the port status, although I know when I busy the ports by using four phones to call into the 4-port system I get a busy. When the VM is locked up, I only get a ring no answer.

I checked the system programming from a phone (I don't trust the PCDBA as it reports different settings than what I programmed by hand) and the settings were the same as the original setup and matched my paper copy. I did a cold boot and reprogrammed the system from scratch.

Now as for the PCDBA reporting the wrong information, the only two areas I've seen it incorrect were:
Ringing assignments - I programmed in by hand five extension - 101, 102, 103, 110, and 111; and those extension correctly rang; but the PCDBA showed 101, 102, 103, 108, and 111. I've played with the programming back and forth many times and have no idea why. When I use the PCDBA to set the ringing and upload it back to the switch, it works fine, but when I download what I just uploaded - 108 reappears.
Extension port swapping - that was a mess and I ended up doing it by hand.

As I recall, it was not the latest PCDBA version for the DX-120.

I'll see what tomorrow brings.

Another glitch in the PCDBA. I used the phone to reassign my attendant phone from port 101 to port 116 and it works just fine.

PCDBA shows port 101 swapped with 114, not 116. Oh well...

OK, looks like I'm going to have to replace the cabinet.

I did a cold boot of the system early this morning, did new programming from scratch (not a system backup) and the VM crashed a few hours later.

Since the original programming work for months, I don't thinks its a programming issue, and am using the same basic parameters on several other DX- VM systems

I loaded the K17 firmware update today and will see what happens.

I did have a chance to look at the system while in failure mode.

1) Keylink showed the ports idle.
2) Dialing each port separately returned a busy siganal.
3) Dialing the UCD group returned a ring-no answer (yes, I check to see if there were any other members in the VM UCD groups and there were only the three VM ports).
4) After I disconnected KeyLink, the VM rebooted itself. I heard it go through the boot-up beeping, which it usually doesn't do.

You have three voice mail ports??? Also I seem to remember that K17 didn't auto reboot the mail after a power restore. Sounds like multiple issues.

Sorry...I meant FOUR VM ports.

Anyway, the last four times this week, there were no power outages.

No go...the VM worked all weekend until noon on Monday. I'm going to replace the CPU board early tomorrow morning.

I'm still puzzled why you can get a busy when you dialed the individual ports, but a no answer when you dial the UCD group number, And I already checked, there are only four stations used fr the UCD members

did you make the ext VM in the extention program

If I didn't, it wouldn't work at all. It works, then crashes, then on at least once occasion started up by itself. I don't see it as a programming issue, as it had been working for several months with a good set of program settings.

I'm going to replace the CPU card.

Any update on this?

The new CPU I replaced about a week ago didn't fix it. I pushed the L18 firmware last Friday and no problems yet. Other than I had to go out and reset the VM Tuesday morning after the power was off for ten hours last Monday afternoon.

I drove up across the street to upgrade another system with L18 and the place was dark inside...I asked if the power was out and she said it was. I thought she was joking...until I walked around the place and in the dark all I heard were UPS systems beeping. She was wasn't joking. It went off minutes before I got there and was out will middle of the night.

The battery back up on the system with the VM problems must have run down and the VM didn't' come back up when the phone system did...that appears to be a common complaint.

THE..THEN when I was there today to push the L18 like I had planned on doing Monday...the lights flickered in teh middle of the upgrade, but everything was on battery backup.

Its been a week now, and now calls of it going down.

But the L18 does have a few quirks...

I'll start making notes.

So what's new about that? And still no documentation on it, I'm guessing. Let us know when you get it put together. I'm very interested to see what you've found.

I'll start a new thread after I confirm it with my system that I'm about to flash with L18.

UPDATED: No need to start a new post. See next message.

To recap, I'm going to refer to two systems.

The first is where the VM kept crashing and upgraded to L18 - so far its been stable after over a week, when the week before it was crashing on a daily basis.

The second system is the one that had the lighting hit. I upgraded it to L18 which I shouldn't have really done as if problems start appearing, I wouldn't know if it was damage related or L18 related.

After the lightning strike, port 157 of the VM was intermiitant as it would not always answer, so I took it out of the VM hunt group and changed the extension type to EXT. He was happy with just three active VM ports.

After the L18 upgrade, the MSG WAIT light lit up, but instead of the display showing, "New VM Msg" (or whatever the display says), it now shows "Msg from 157 > Reply/Del"

Furthermore, the DTMF tones are lagging after you press a button. It was so noticeable, it indeed caught my attention. Customer also noticed it.

I just flashed my personal DX120 with L18 and I cannot duplicate the two above problems. My msg waiting display still says the same as before and the DTMF is still pretty fast. I'm going to use the Feature 1# ("TOUCH TONE ON") to turn on the pacifier tone and mask the actual DTMF.

I really think he should replace that system.

Anyway...thanks for all your input. The DX120 is a robust system with low cost scalability and cost effective over all installed sizes.

Update on crashing VM.

After three weeks it went down today.

The first thing I did was connected to it with KeyLink and it showed all four ports idle. The activity graph showed no activity past the time they called me.

Dialing each extension number, 159 ~ 162 gave me a busy.

Dialing the hunt group 433 gave me ring no-answer.

I verified I still had the replacement CPU broad installed.

I'll replace the VM card if it goes down again.

We just went back out on one today, similar symptoms. It seemed to have happened after it rained. In the past, it was a 4+4 flash voicemail on a DX80. Upgraded as high as we could, and eventually swapped out the DX80 for a DX120. Problem all but went away, until today. Ports all show idle, but are either busy or RNA when called (I think mostly always RNA). This started many months ago. Lines in that area are pretty bad, often having problems when it rains. Not sure if there's any correlation or not. We thought it had to do with the fact that we were using the UCD with prompts situation (I probably started a thread on it long ago), but maybe it's been weather-related all along. :shrug:

Update: Here is the thread I was referring to. It looks like, based on my previous description, it was busy when called, even though all ports showing IDLE. Sounds like maybe the same situation that you're running into.

Justin

To recap, the VM had been running fine for about three months and then suddenly started having the problem. And the problem got progressively worse from one reboot a week until the third week and it was locking up daily.

Comdial had me turn of the automatic 3 AM restart and made sure all my dial codes pointed to valid prompts. The VM was stable for the following three weeks. Unfortunately I replaced the CPU at the same so I didn't know what fixed the problem. However it locked up yesterday. I'll wait a few days to see if it starts getting worse again.

Lines are T1, so not likely a bad line issue, per se. The power does go out for a second at least once a week, but the system is on a UPS.

I tried both ringing to four specific phones and then a UCD group and made to difference.

UPDATE - -

It was running great for about two months after I upgraded to F17, and turned off the 3 AM reboot.

BUT...It crashed four weeks ago, last week, and again this week - TWICE on the same day. Thats four times this month.

I'll upgrade to F19, but I still suspect a motherboard / power supply issue as I've already replaced the flash card VM and CPU board.

F19 info says it does fix the problem of VM not coming up after a power outtage, but I have this system on a battery backup and customer says there has not been any significant power outtage (it does go out for a few minutes almost every week though and never on the days that the VM has been crashing) and another customer right next door never has a problem with an identical system.

A Vertical rep told me "I'd reboot the entire system instead of rebooting the VM only when this happens."

Don't ya love that!
I always tell them that a reboot may mask the symptom, but it doesn't fix what ever is causing the trouble.
Mark

Actually what eh said in full quote was, "I didn't say this. I'd reboot the entire system instead of rebooting the VM only when this happens."

In regards to VM locking up,

I had a DX120 with 7271C VM doing the same thing

After 2-3 weeks of reseting system I made some changes in the phone system UCD groups for routing calls from voicemail.

Make sure if the voicemail routes to a UCD group, that the (reroute destination) isnt an extension that is located in the that UCDs (members list).

That fixed that problem.
Now the problem with the time being incorrect on voicemail has started (4 months after) I think I will update to L19 software and see if this takes care of the probem

Thanks
new to forum

I recently had this trouble on a DX80 system 18 mths old, first it had a bad flash card which I replaced. Then it started shutting off every day, a reset would bring it back but it would shutdown again in a day or less. I replaced the board & flash and had the same trouble. Yesterday I did a cold start and reloaded the database from a copy I had from several months ago. So far it's still running, it's been 24 hrs.
Mark

Mine is down again this morning, appears the new VM board has a corrupted flash card, I'm preparing another new one for them now. I'll follow up tomorrow.
Mark

Don't ya just love it. You make maybe 1,500 profit on the sale and then piss it all away on go backs.
Have you tried grounding the spare pairs? JK. That was always my favorite response from Comdial tech support in the 80's.

Bad flash card. check disc found numerous problems, running the fix appears to work but once the system use hit the bad sectors the allocation errors returned. Disc doctor did not work and wiping it clean and reloading had exactly the same problems. I replaced it today after testing it here for hours and bombing it with calls. Seems ok now, we will see. You can't get an rma just for the flash card, you have to return the whole thing, so that's what I'm doing. Pity the dealer that doesn't keep stock of their products.
Mark

I upgraded to last week and the system is still going down every few days. The only thing I haven't replaced yet is the cabinet and the flash card.

I can record all the prompts and messages on to a CD and pass out individual CDs for about ten users.

Did you run the check disc and NDD on the flash drive?

There is something you could keep in mind, and it "typically" works. Try a double cold start.
It sounds crazy but give it a try.
Remove your vmail card and move the cold start jumper to the cold start position and wait 10 minutes. The full ten minutes.
Move the cold start jumper to the normal position and power the system up.
After it is powered completely up, power the system off.
Install the voicemail card and power system up. After it is powered up try calling the voice mail. It may or may not work.
Power it back off and Install the cold start jumper. Wait Ten Minutes
Remove Cold boot Jumper. And Power up system.
And I would rebuild your program all over again.
Keep in mind never have you laptop connected to the system when cold starting the system. I have notice that causes problems too.
And if you keep having problems remove all the station and co line from the system and try these steps over again.

Remember patiences

Those are all good troubleshooting steps, but you can save a lot of that time by checking the flash card with check disk and NDD. If the trouble occurs with the flash mail after the mail system has booted, it's likely to be bad sectors on the flash card.
All the resets and cold starts won't help to resolve damaged sectors of a flash drive. In my instance, check disk and fix disk cleared the trouble, but only until the system needed to use those sectors again, then it failed again. Even though the VM is "in skin" the phone system programming is not likely to cause these type of troubles. They really are still seperate entities even if they are "in skin" VM systems.

I usually do a cold start when I add or remove boards, the manual doesn't say you have to do it...but it does resolve some weirdness.

Anyway...the system has been working fine up to four weeks ago and now needs a weekly reset. I looked at the routing tables last week and noticed the "press 2" option from the 800 box pointed to a 888, which I know perfectly well it was pointing to 812 in the past.

Knowing that pointing to wrong places can cause some problems, I reset it to 812 and told the customer it *may* fix the problem, but I really think it was the flash card and it would crash again. I didn't want to try to "fixes" at once as I wouldn't know which fix fixed the problem. They were cool with that. At least I set their expectations on what I was doing to resolve the issue.

Yup...it went down again today. Thanks Mark for mentioned a replacement flash card...which was the ONE thing I haven't replaced yet.

I can record all the prompts and user messages onto my laptop and then record them back on the new card as a new message.

Hope it all works out.