Latest beta firmware breaks comms with TA

[jziggity]

A couple of days ago I upgraded my InfiniTV4 PCI firmware to the latest beta version to see if it would help with the audio clipping/video freezing issue I was having on one channel in particular. It did not, and has since introduced a new, much MUCH worse problem. Every 24-36 hours, my InfiniTV card now loses communication with my TA. Any channel I try to tune spins the media center circle for about 1 minute, then returns an error "This channel is temporarily unavailable. Try again later. (SDV error: 1)." During this time, the TA light (Cisco STA1520) is on steady, indicating proper operation. The web address for the PCI card (192.168.1.200) indicates all is well with the TA, EXCEPT it will not bring up the TA's diagnostic page. When I click on that link, it states "Request timed out. Try again later."

I'm no detective, but I'm going to say that is the culprit right there - the InfiniTV card is no longer communicating property with the TA. Resolution thus far has been to power cycle the TA, then keep clicking refresh multiple times (upwards of a dozen) on the webpage for the TA diagnostics, until it finally starts communicating again. Even when the diagnostics page comes up, sometimes it doesn't stay for long before the "Request timed out" error message comes back again, so I have to keep refreshing until it returns the appropriate page consistently. In the last 3 days I have had to go through this process twice. Not surprisingly, recordings fail when the TA/InfiniTV communications are down.

I'm betting all my chips on the beta firmware upgrade causing this, as I haven't made any other system changes. I have submitted a Ceton support ticket with all of the information and requested they provide me instructions on how to roll back my firmware. If they respond promptly with said instructions, I am hopeful that I will be able to take the InfiniTV back to a (mostly) stable state once again. If not, I'll be pulling the trigger on the HDHomeRun Prime. I am so sick of babying this card. Screw the WAF, what about the MAF (MY approval factor)?

[JohnW248]

Next time you see a SDV1 error try the following in this order:
1. unplug the usb connection and then re-plug the usb connection--see if it tunes
2. Power cycle the TA--see if it tunes.

If you can open the TA diagnostic map in a web browser (and leave it open) see if you can refresh that browser page or if you get TA timed out, check the web gui and click on the TA tab and see if it says disconnected or disabled.

Do a diagnostic scan when you lose the TA and another after you get it back up. Send both to Ceton. Don't know if you have a cisco or moto system but they handle things differently. Your audio problem (was it ever resolved? I asked some questions about your set-up and I don't know if you ever answered) might have been incoming from the service provider. When only one channel demonstrates a problem, it is more common a source issue and not a media center, extender issue.

[jziggity]

I was very pleased to see Ceton answered my trouble ticket within 40 minutes, and gave me the link and instructions on how to roll back my firmware. I was half expecting a laundry list of things to try from them before acquiescing to the rollback, but they did not. I'm just going to do that (probably tomorrow), and see if the problem persists. If so, I will continue with JohnW248's suggestions.

John, I did respond with some additional setup info on the other thread about the audio issue. I haven't been watching that channel in the last couple of days, so I'm not sure what the current status is. I am very much inclined to believe it is an issue from TWC, but the challenge will be proving that, especially to them. If a customer has any non-standard (read: not theirs) equipment, it must always be the equipment's fault first, until proven otherwise.

[JohnW248]

You know, I've never had that problem with TWC. I imagine there are a few CSRs that'll try that but over the years I've gained a reputation in my district and no longer do they automatically want to schedule a truck roll when I call and tell them there is an issue connecting to the node or high davic error counts, etc. Over the years I've had every repair tech, lead and forman here at one time, the first cableCARD installation resulted in four or five (I've forgotten) techs here at the same time. They kept calling other guys to see what was going on (and that was with Vista and the ATI DCT). They've replaced everything I"ve asked including the drop to the house, etc.

So you can get results from them, document what's going on. I've printed out pages from the cableCARD and TA diagnostics (which are all their equipment) and handed it to the guys when they're here and showed them what it looked like when it was working. So keep at it, a single channel issue is their problem, not yours.

[Embiggens]

I hope I'm not threadjacking but what do people consider an 'acceptable' frequency to lose tuning adapter connectivity? I seem to need to power-cycle my TA once or twice a month. Is this something that should *never* happen? If so, yeah, I will also take your advice and record some diagnostics and call up my provider probably.

[JohnW248]

I hope I'm not threadjacking but what do people consider an 'acceptable' frequency to lose tuning adapter connectivity? I seem to need to power-cycle my TA once or twice a month. Is this something that should *never* happen? If so, yeah, I will also take your advice and record some diagnostics and call up my provider probably.

It seems completely random that one will lock up. I have six 1520s and no real pattern to when one might lock up and when it happens all the others are working just fine. Never have been able to pin what causes it but I keep a page open to the TA diagnostic map on five on them (the last one is on a Tivo that doesn't get looked at much anymore).

[jziggity]

I completed the firmware rollback earlier today, and the process itself went fine. After a reboot, there was still a hiccup with the TA like I had before, "Request timed out, try again later" when I attempted to access the TA diagnostics page via the Ceton device webpage. That was not encouraging. After multiple refreshes, it once again came back up and stayed that way.

The true test will be about 24-36 hours from now. That's about the time frame for this issue to repeat itself. If it happens again, this Ceton card is going to be put out to pasture. I'm just done dealing with it. First $100 will take it, after an HDHomeRun is installed and running that is. I know someone is going to say, "Well if your TA is causing the problem, it won't be any different with an HDHomeRun and the same TA." Maybe, maybe not. My plan is to order a new cablecard and TA and start over fresh. Stay tuned... *drum fill*

[fiteclub]

I had this exact same issue about a month ago. Same symptoms as jziggity. Ceton had me roll back to 1.2.2.6. This solved the issue but I lost the ability to disable EAS. Oh well, hopefully Ceton will fix in the next firmware.

[jziggity]

$^#%@$&!!! So, just got a message from my wife that the TV crapped out again. I logged in remotely (not home right now) and sure as heck, the TA communications is down, and I can't get anything on any of the tuners. Even though everything inside the Ceton webpage is reading normal, this is what I'm getting in the Ceton diagnostics program.

How is this even friggin' possible?? How can it be showing that "A valid CableCard was not detected..." while at the same time showing "CableCard Validated" and "CP Authorization Received" and "OOB Locked" and "Receiving OOB Messages." This makes absolutely no sense.

Thus begins the conversation, "Do we even need cable anymore...?" NOPE!

[jziggity]

Ugh, might have hit the panic button too soon, not sure just yet. I forgot about a brief storm-related power outage at home about 4.5 hours ago. When I got home, everything tuner related was still down. I went and looked at the actual TA itself, and it was showing 6 blinks. Another reboot of the HTPC, and everything is up and running again. My Ceton Diagnostics program still looks like the one above though, which makes me really nervous that it will fail again in the near future.

I guess the 24-36 hour clock starts over again.

[JohnW248]

Here are two "unsupported" tips to keep from rebooting your computer. When you have this problem but CAN get to the web gui, do the following in this order to restore operation.

1. Restart the cableCARD which will re-detect and bind. To do this go to the cableCARD tab in the web gui and scroll down the page and click on "CLEAR CHANNEL MAP" This will restart the cableCARD.

2. If the TA is blinking, power cycle the TA

3. Final step before rebooting. Go to the web gui, click on LOG and at the bottom type in REBOOT and hit enter. The card itself will restart along with the cableCARD. This should have the same effect of rebooting.

If these don't restart the card, then something in Media Center failed.

One final trick: If the card shows everything working but you can't see the tuner in Media Center ("No Tuner Available, if using a STB recheck IR"), you can go to detection again. You can either run this from an elevanted prompt or from the tasks. The tasks method is easier for me to explain, right click on computer and choose manage, then open task library, keep opening to media center, find OCUR discovery and click once and then from ACTION drop down and choose RUN.

Or you can run this from an elevated dos (they call them command nowadays) %windir%\ehome\ehPrivJob.exe /OCURDiscovery

[jziggity]

Thanks for those tidbits John. The TA communication was lost again just a little while ago. It was corrected quickly after I did a refresh of the TA diagnostics page again. Unfortunately all of this fiddle-f'ing around is just not worth it anymore. The card should not need constant user interaction on a daily basis to stay running. It should just work. As a last ditch effort, I did a system restore to a time when I believe all was working fairly well. The last time I had an unresolvable issue with the Ceton card this worked, so here's hoping. I will remain cautiously optimistic, but if this doesn't work it is either time for the HDHomeRun, or an antenna setup.

[jziggity]

Predictably, my Ceton card has lost communications once again with the TA, at about the 24 hour mark. I can no longer continue troubleshootING this thing. My family and I have lost patience with it. I plan on purchasing an HDHomeRun Prime and staring over fresh with a new CableCard and TA from TWC.

Thank you John for all of your assistance. I thought you might find this part interesting - this morning, after the most recent comm failure, I skipped to the end of your list and just went with the "reboot" option on the log page of the web GUI. I literally did NOTHING else - no reboot of the PC, no power cycle of the TA, nothing. I watched the light on the Ceton card blink during the reboot process. As soon as that was finished, everything immediately started working again. This to me has to point to either 1.) a bad CableCard that now requires constant rebooting on a daily basis to maintain communications with the TA, or 2.) a bad Ceton card that requires daily rebooting. Since I didn't reboot the TA or reboot my computer, or change ANYTHING else, this clearly has to point to one of these two issues.

I responded to my original Ceton trouble ticket and asked them to take my card back as an RMA in exchange for a new one. I highly, HIGHLY doubt they will agree to this, but I told them I can no longer devote any more time to this thing. Something clearly broke during the last beta firmware update that I did, and rolling it back again did not fix the issue. If Ceton says no dice, or wants me to jump through another 12 hoops of various things to try, I will just get the HDHR and put my Ceton card up for sale. Perhaps it will work better for someone else with a different CableCard or for someone that doesn't use a TA.

[JohnW248]

Unfortunately, a reboot of the tuner card does too many things to provide a pinpoint on the problem. The card restarts and reloads its FW, the cableCARD re-binds and validates, the tuner and cableCARD bind with the TA and a new channel map is downloaded by the TA. Next time, if you are just getting SDV1 errors try unplugging and replugging the USB connection from the TA. That will normally cause the TA to re-establish communication with the host and download a new channel map and it is a "fast" recovery in most cases. If the TA does go through all it's flashing light routine, then it has also lost upstream or downstream and needs to re-authorize and get an IP from the node.

You might find that this is a problem with the cable company and not your equipment. If you have a Cisco TA, go to the diagnostic map, open RF Statistics and go to page two which should look like this:
S/N: 39 dB
Seconds: 9065
Corr Bytes: 0
Uncor Blks: 0
Errs Avg/Inst: 0 / 0
EQ Gain: 1.0
CURRENT RDC
Freq: 15.000 MHz
Power: 41 dBmV -- this is the upstream power used to connect to the node. Anything above 50 dBmV should be investigated as it points to a problem between the TA & Node
Delay: 638 uSec ---this is a timing offset so all the TAs don't collide at the node each TA will have a different delay
Retrans: 0 ---numbers here show multiple calls from the TA to the node for channel information
NextPage Previous Refresh Home

[jziggity]

Ok, it took quite awhile to be able to get to that screen thanks to the comms going down yet again, this time within only a few hours. I tried the USB cable unplug/plug back in as you suggested. The TA immediately began its cycle of 6 blinks, and then would not recover after the USB cable was plugged back in. I then went back to your previous post and did the "Clear channel map" for the CableCard, which really fubarred everything. It took quite a bit for a new channel map to be downloaded, and the TA would not stop its 6 blink cycle until I sent the "reboot" command via the log page again. I actually had to do this twice. Now it is running again, although not for long I'm sure. Here's the data I got from the TA page you referenced:

S/N: 36 dB
Seconds: 16974
Corr Bytes: 0
Uncor Blks: 0
Errs Avg/Inst: 0 / 0
EQ Gain: 0.9
CURRENT RDC
Freq: 16.500 MHz
Power: 37 dBmV
Delay: 652 uSec
Retrans: 64

Everything looks about the same, with the obvious exception of the "Retrans:" field. As I am typing this, I lost TA comms yet again.

Since the only consistent solution appears to be a full Ceton card reboot, doesn't that still point to the Ceton card as being the problem? I know you said there are multiple things going on during the reboot, but I think I effectively eliminated the CableCard as the issue by trying the "Clear channel map" option, which didn't solve the issue.

I'm also on the line with TWC right now ordering another TA and CableCard. I will most likely try the new equipment in the current Ceton setup before deploying it in a new HDHR, just to satisfy my curiosity. This TWC phone call is quite humorous - the CSR just keeps saying "Ok, ok..." over and over again as she struggles through the request for new equipment. She has no friggin' clue...

[JohnW248]

It looks to me that your problem is a communication issue between the TA and your Node. The problem would very likely be the Node itself or some "intruder" on the drop or tap that is causing trouble. Rather than ordering equipment, get someone at a higher level that can look at your TA. They cannot see cableCARDS, but they CAN look at TAs. Under Version & MACs are the mac address for your TA, under Switched Digital Video is your IP address of your TA assigned by the node. Either the CableCARD help desk or level 3 cable TV can look at your TA and once they see that 64 (which occurred since your last TA reboot BTW) they should send out someone to investigate what your issue actually is. Based on this I would expect a lot of SDV1 errors and your problem with the cableCARD/tuner could easily be a time-out problem due to the high number of Retrans. But get the experts involved and believe me TWC does have experts on SDV....been there, spoken to them.....

[STC]

John wow, you really do know this area. I am impressed with your technical prowess.
Fancy becoming a resident TGB Cable Guru?

/oh hold up....you already are lol.

Well, thanks! Keep up the good work!

[JohnW248]

John wow, you really do know this area. I am impressed with your technical prowess.
Fancy becoming a resident TGB Cable Guru? 8-)

/oh hold up....you already are lol.

Well, thanks! Keep up the good work! :D

Would I get a badge and compression crimper?

[jziggity]

On the line now with cablecard support desk. I know it sounds like I keep bashing TWC, and with good reason - that's based on my experience with them. It is notoriously difficult to get a hold of someone in my area who knows what they are doing. It's a "no one sees the wizard" type of mentality. However... today I guess I got lucky, I'm talking with someone who actually knows about these numbers apparently. I know John will be shocked - they are saying my signal levels are low and the communications with the TA keeps going in and out.

In particular, the cablecard desk said the numbers under the RF Statistics screen that they are looking for are: CURRENT FDC Level to be 5, I'm at 0. They also want the Current RDC Power (page 2) to be near 51, and I'm at 37. John, does this all sound right?

And just to rub salt in the wound, the cablecard desk person I'm talking with says that she sees an order for a DTA on my account. Friggin' CSR morons got it wrong again. Even if I sent them a picture of what I want, they would still get it wrong. No one listens anymore, company here excluded of course...

Truck roll scheduled for tomorrow afternoon - here's hoping!

[JohnW248]

Are you talking to Barbara? Hope so, she's good and she'll get people out there. 51 is within spec and it probably means you're further from your node than I am. But in the future when you have trouble, it's best to go directly to the cableCARD help desk--it'll save you a lot of aggravation...

Looking forward to an update after your truck roll. You never know what they might find doing a visual on your line either. I once had a splitter with another drop running off mine. When the drop was replaced the tech just cut it figuring if it was legal they'd get a service call---never did so it was a midnight installation. But those sort of things can really kill your signal level. Also the tech can do a full spectrum scan of your incoming signal and decide if a two way amp is necessary at your location.