Mailbox fails 1

[davea2]

Hi

We have a ticket open with Avaya for this but thought I would ask here also

We have just deployed a cloud SE (AWS from the correct community AMI on IPO-CPE-11.0.4000.74)

All is working well we have an additional HDD for Media Manager also.

They have a group mailbox for taking orders overnight and this is failing once a large number of messages have been left.
Typicall it will fall over at somewhere between 120 and 200 messages.

If we delete or reset the mailbox and restart the vmpro service all is good again until the next time the messages reach a high number.

The reson I thought I would ask here is that Avaya, as usual, want all sorts of traces when it is failing which is fair enough but the only way to do this is to allow the system to fails when taking orders, thus losing te company much money in lost orders so if we can gain some insight binto this without having to lose messages it would be better.

At present we have a workaround in place where night calls get diverted to a mailbox on our system here!

Any thoughts/ideas appreciated!

Cheers

Dave

 

[Okkie26]

Ia there enough harddisk space? Since VM pro does not use the extra HD for voicemail messages. (as far as I know).

 

[davea2]

Yes I think there is plenty of space.
This is the current state:-

Filesystem Size Used Avail Use% Mounted on
/dev/xvda1 99G 13G 81G 14% /
tmpfs 1.9G 4.0K 1.9G 1% /dev/shm
/dev/xvdf1 470G 805M 445G 1% /additional-hdd#1/partition1

 

[IPOLackey]

Hi Dave,

I've seen this also but haven't replicated yet as our work-around was to enable VM to EM and just forward all the messages.

Only other way is to create a new HG and manually leave 200 messages. I might test on our lab instead which is on the latest SP and see what happens.

 

[davea2]

Having had a poke around some other threads and looking at the files on the server, I have come across this settings

MailBoxQuota=30

In vmpro_settings.ini

The meaning I think is that each Mailbox has a max limit of 30Mb, so approx 30 minutes which is nowhere near enough.
Their old system was on 8.1 I believe which did not have this setting.

We are going to try changing this parameter on our test kit to see what happens.

Surprised that Avaya did not think of this right away if indeed it is the problem...

 

[sizbut]

The format for recordings was changed a short while back so 30MB per mailbox is now approximately 60 minutes.

That said, 200 customers and 60 minutes is 20 seconds per customer (and that 200 is when the system fails, so presumably they have more than 200).

Even if they had an agent there overnight that agent would be handling a customer call every 5 minutes minimum. Personally, if the business is really so important to them, pay for some staff to handle it as it happens rather than payign staff to handle it 12 hours later.

Meanwhile, its Voicemail Pro, use a callflow to do something out of the box like put each hours' messages in a different group mailbox or to distribute each to a different individual agent mailboxes in turn (probablhelp with the agent processing the next day as each agent will have their share of the night's calls) or ...

What would be their expected number of messages overnight if things were working okay?


Stuck in a never ending cycle of file copying.

 

[davea2]

Well it's deffo the quota:-

29/12 14:40:45.149 vmpro (07,5) 77d1,05d55b70,19527: Mailbox Telesales is full. Size - 31 MB. Quota - 30 MB
29/12 14:40:45.149 vmpro (07,5) 77d1,05d55b70,19527: MailBox Recording cannot continue as Mailbox Telesales is full
29/12 14:40:45.149 vmpro (09,2) 77d1,05d55b70,19527: Session: 00003db8 - Recorder - not enough disk space to perform recording, aborting record for /opt/vmpro/Accounts/Telesales/MSG00161.wav
29/12 14:40:45.155 vmpro (23,5) 77d1,05d55b70,19527: SoundsManager::ConvertWAV(0xf73b4c18, "/opt/vmpro/Wavs/en/svm_04.wav" [aLaw], 0xf73c3388) - Took 3 ms to process, skipped 0 samples of silence (0 at beginning, 0 at end), return is 34080
29/12 14:40:45.155 vmpro (23,5) 77d1,05d55b70,19527: Succeeded in loading sound bite "/opt/vmpro/Wavs/en/svm_04.wav" [aLaw]


With this particular Customer, getting them to change their MO woud be nect to impossible.
On 8.1 with their old C110 card there was no issue, and my guess is they will expect no issue with a newer system.

We tried setting the quota to 5MB on our test system but didn't seem to have any effect, we could still leave messages at 8MB

There must be a way to change this surely?

 

[davea2]

about 200 messages would probably be the norm

 

[joe2938]

As stated earlier, I would go the route of voicemail to email, you could create a email group list so several users could handle the large amount of emails.

 

[janni78]

You could try and change the setting MailBoxQuota=30 in /etc/vmpro_settings.ini

"Trying is the first step to failure..." - Homer

 

[rdoubrava]

I had a similar problem on our system.
I have increase the number to 50 on our system.
You have to restart the vmpro service.

Avaya doesn't recommend increasing the number.
I haven't had any issues with using 50.

 

[janni78]

Maybe I should have looked through the thread as you already found it =)

"Trying is the first step to failure..." - Homer

 

[davea2]

Changing the parameter on the site system seems to do the job.
It appears we do not need to restart the vmpro service either.

We will also have to change housekeeping to get rid of deleted mails sooner to keep space down.

This seems to be a very low limit for a system like the IPO, especially if Avaya do not recommend changing the parameter.