Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Reese Boisse]

Hi Everybody. Haven't been here in a while, but I'm having a problem that's exhausted pretty much all other resources.

I'm using a system of 2 AudiaFLEX chassis, which control audio for a 3-into-1 combinable training/conference room. Each room has three Shure ULX wireless mics (one lav, two HH) and two Clearone ceiling mic arrays for room pickup. All mics go into one automixer per room. The rooms combine using matrix mixer and AM combiner presets recalled via Crestron AV2, and it uses two VOIP-2 cards. During events I usually control these over Wifi which I know is not ideal, but is my only easy option.

So, the problem: Every once in a while, after the wireless mics have been used for a little while, the matrix mixer seems to stop or intermittently send to certain outputs. So, speaker is at the podium, everything sounds great in the room. But people on the phone hear the speaker cutting in and out, or nothing at all, and our recorder, which piggybacks a VTC codec to capture A/V, receives this audio as well so we are able to hear the problem recorded, after the fact. When I'm able to stay connected to Audia SW (getting to this...) this is reflected in the meters; mic inputs flash, room outputs flash, VoIP/VTC outputs hover around the noise floor. The room mics also fail to go out over VOIP and VTC if they are unmuted.

Each time this happens, it's either preceded by a brief SW hang, a loss of connection to the system, Error 1a03 "when trying to associate a communications channel with a device session,” and timeout errors when I try to reconnect, OR, I'll remain connected and able to still control certain objects like level controls, but, for instance, opening a control dialog for an auto mixer will cause the SW to hang until I close the dialog window, and clicking any given button may or may not have any effect.

These are connected to a small-ish, but IT-only and well-managed subnet of a larger network, and Biamp's (excellent) support has so far pointed to indeterminate network problems as the cause of this, as the Audia ethernet control ports are only 10MB half-duplex (which I found weird.) As one troubleshooting step recommended to me, I moved these off of the primary subnet which had significant broadcast traffic to their current location where there is very little, but the problem persists.

I've also put a PC running Docklight as a serial pass-through between Crestron and Audia to see if Crestron is spitting out something weird that can't be dealt with, but I'm missing something as 5 serial cables which all pass a loopback test won't even show me what Crestron is sending for more than one line, and getting that all the way to Audia is proving beyond me.

My remaining steps include setting up a PC on a mirrored switchport to just run Wireshark for the duration of every event until it fails, setting up a discrete network just for the Audia with a little wifi router for my connection, lobbying for a new port in the wall that could go to a hard coded 10M H-D switchport, rather than sharing a Gb port with a couple other light-usage control interfaces behind an unmanaged switch, and going to work at Pull-A-Part.

Sorry for the wall of text, but this seems to be an esoteric problem. Been in the stages of trying to figure out where to begin troubleshooting for too long now, so any and all input is appreciated.

Thanks,
-R

[David Sturzenbecher]

Hi Everybody. Haven't been here in a while, but I'm having a problem that's exhausted pretty much all other resources.

I'm using a system of 2 AudiaFLEX chassis, which control audio for a 3-into-1 combinable training/conference room. Each room has three Shure ULX wireless mics (one lav, two HH) and two Clearone ceiling mic arrays for room pickup. All mics go into one automixer per room. The rooms combine using matrix mixer and AM combiner presets recalled via Crestron AV2, and it uses two VOIP-2 cards. During events I usually control these over Wifi which I know is not ideal, but is my only easy option.

So, the problem: Every once in a while, after the wireless mics have been used for a little while, the matrix mixer seems to stop or intermittently send to certain outputs. So, speaker is at the podium, everything sounds great in the room. But people on the phone hear the speaker cutting in and out, or nothing at all, and our recorder, which piggybacks a VTC codec to capture A/V, receives this audio as well so we are able to hear the problem recorded, after the fact. When I'm able to stay connected to Audia SW (getting to this...) this is reflected in the meters; mic inputs flash, room outputs flash, VoIP/VTC outputs hover around the noise floor. The room mics also fail to go out over VOIP and VTC if they are unmuted.

Each time this happens, it's either preceded by a brief SW hang, a loss of connection to the system, Error 1a03 "when trying to associate a communications channel with a device session,” and timeout errors when I try to reconnect, OR, I'll remain connected and able to still control certain objects like level controls, but, for instance, opening a control dialog for an auto mixer will cause the SW to hang until I close the dialog window, and clicking any given button may or may not have any effect.

These are connected to a small-ish, but IT-only and well-managed subnet of a larger network, and Biamp's (excellent) support has so far pointed to indeterminate network problems as the cause of this, as the Audia ethernet control ports are only 10MB half-duplex (which I found weird.) As one troubleshooting step recommended to me, I moved these off of the primary subnet which had significant broadcast traffic to their current location where there is very little, but the problem persists.

I've also put a PC running Docklight as a serial pass-through between Crestron and Audia to see if Crestron is spitting out something weird that can't be dealt with, but I'm missing something as 5 serial cables which all pass a loopback test won't even show me what Crestron is sending for more than one line, and getting that all the way to Audia is proving beyond me.

My remaining steps include setting up a PC on a mirrored switchport to just run Wireshark for the duration of every event until it fails, setting up a discrete network just for the Audia with a little wifi router for my connection, lobbying for a new port in the wall that could go to a hard coded 10M H-D switchport, rather than sharing a Gb port with a couple other light-usage control interfaces behind an unmanaged switch, and going to work at Pull-A-Part.

Sorry for the wall of text, but this seems to be an esoteric problem. Been in the stages of trying to figure out where to begin troubleshooting for too long now, so any and all input is appreciated.

Thanks,
-R

Have you called Biamp? Their support is top notch.

[Reese Boisse]

Have you called Biamp? Their support is top notch.

Thanks for the response. I've called them a few times and the most we can work out is that it seems some external (probably network-related) factors are causing wildly inconsistent behavior. Kudos to them though for sitting on the phone with me for quite some time to get here.

This system used to be similar but simplified, with Nexias in place of Audia and no room mics. We did see the same problem connecting to the control interface with error 1a03, but audio was unaffected at the time so it wasn't as pressing an issue. Support at the time recommended checking Nexlink connections, so we replaced those with new shielded Cat5e but still the same problem.

I can define a couple somewhat specific questions:

1. Just how much broadcast traffic is too much when dealing with AudiaFLEX on a network? The amount it's receiving would be inconsequential if it were any other device, but I worry that maybe we're overloading that half duplex connection if that could cause varied problems?

2. Is it possible for some sort of bad RS232 data to break the TCP connection to a system and cause crazy behavior? If I recall, Biamp said this was unlikely but possible.

Thanks again, even this is more feedback than I got in months on their old user forum.

[Reese Boisse]

Update, spoke to Ben Wilmot @ Biamp support who recognized this as meeting all criteria to be a potential instance of a rarely-seen bug in .850 firmware. We may be running into an issue where the open mic limits of auto mixers and auto mixer combiners conflict when a combiner is in a particular state (though I'm forgetting specifics,) causing the auto mixer combiner to lock up and as a result anything that looks to the combiner will fail.

One thing I neglected to mention in the OP which was a tipoff was error 1902 coming after 1a03 when continually trying to reconnect.

If this space remains quiet, then we can say a firmware update to .870 fixes this. If not, isolating from as much broadcast traffic as possible will be next step.

Thanks to Ben @ Biamp!

[ProSoundWeb Community]