InfiniTV reboot issue

Ask fellow members about Ceton's infiniTV tuners here.
Forum rules
Ceton no longer participate in this forum. Official support may still be handled via the Ceton Ticket system.
joevulture

Posts: 22
Joined: Sun Jan 01, 2012 3:48 am
Location: Indianapolis, IN

HTPC Specs: Show details

#41

Post by joevulture » Sat Feb 25, 2012 1:22 pm

I'm pretty certain that there is something with the TV feeds that the Ceton firmware doesn't handle correctly that is causing the issue.

I'm using Brighthouse cable, and they mark everything except the locals as copy-once; because of this, almost every channel exhibits the random strip of garbage quite frequently; the local channels do not have this issue. This is an acknowledged firmware bug that I hope they fix soon. It started happening sometime last spring after a firmware update to my PCIe card and my USB Ceton shipped this way from the factory. I just assumed that it was something that got messed up in my wiring or that I had damaged my PCIe Ceton due to excessive heat, despite excellent signal levels and a solid SNR.

I've also got the USB reboot issue, although I don't see it too often, but when it happens while I'm watching live TV, it is annoying. Hopefully the new firmware helps out with this.

Ceton support to their credit has been reasonably easy to work with although they haven't been able to solve my problems (they are working with Brighthouse to resolve a SDV issue, which is my main one at the moment). I'm pretty sure the SDV issue is the fault of the cable company, but again, it's really annoying.

I really don't want to give up the fourth tuner, but a HDHomeRun Prime might be in my future.

That's my rant,
Joe

brainwarped

Posts: 10
Joined: Tue Feb 28, 2012 1:35 am
Location:

HTPC Specs: Show details

#42

Post by brainwarped » Thu Mar 01, 2012 5:34 pm

I was running OK on 1.0.9.5 and then started having a couple crashes. On 2/17 in the crash log there was a segfault so I opened a support ticket.
[42949378.020000] ctn91xx_interrupt_rpc_recvd:79 : (0) queueing rpc recvd: payload_start: 1 msg_type: 8
[42949378.020000] ctn91xx_interrupt_rpc_recvd:79 : (0) queueing rpc recvd: payload_start: 1 msg_type: 8
[42949382.100000] ctn91xx_handle_delayed_event:297 : (0) event card insert mcard
[42949382.100000] ctn91xx_power_setting:106 : (0) manual pcmcia power setting 05a0
[ 400.930000] Fault whee upnp_ocur_serve [1274]: segfaults at 00000000 pc=00070ac4
Ceton had me enable more logging and had another crash on 2/20:
[229644.760000] 1206 total pagecache pages
[229644.760000] Free swap: 0kB
[229644.760000] 7569 pages of RAM
[229644.760000] 166 free pages
[229644.760000] Out of memory: kill process 9646 (upnp_ocur_serve) score 6285 or a child
[229644.760000] Killed process 9646 (upnp_ocur_serve)
redirected stderr
Then they had me update the firmware to 1.1.2.4 and the problems seemed much worse. I kept uploading diags with every crash and then I got a message
We are looking into this issue. At this time, there is no need to collect additional diagnostic captures
I had not heard anything in a couple days so I contacted support about downgrading back to 1.0.9.5 to try to gain some stability and I asked support if they had an ETA on a fix and got this response.
We don't have an exact date available, but we expect to communicate additional information very soon.
So it appears Ceton is acknowledging there is a problem but does not have a fix. I am also hoping that this announcement is not some type of product discontinuation to push their new products, especially since these tuners are now going for half the original price. I am really happy with this product and 7MC when everything is working so I hope they get this resolved.

User avatar
WhatHappend

Posts: 69
Joined: Mon Jun 06, 2011 8:00 pm
Location:

HTPC Specs: Show details

#43

Post by WhatHappend » Wed Mar 07, 2012 3:29 am

Ceton has new beta firmware that seems to be very stable for my set-up so far. Th previous 1.1.2.X betas were unstable for my set-up.

The new stable beta is: ceton_infinitv_beta_fw_20120304_1_1_5_2

Good news is this version seems to fix the "Fault whee upnp_ocur_serve [29]: segfaults at 00000000 pc=0010c468" type crashes. Time will tell if they are totally fixed.

KHTPC75

Posts: 47
Joined: Sat Feb 25, 2012 12:56 am
Location:

HTPC Specs: Show details

#44

Post by KHTPC75 » Wed Mar 07, 2012 3:54 am

I decided to play around with linux end of last month and noticed they had added TA support which required a beta firmware 1.1.3.9. I tried 1.1.3.9 and I could tell from the logging that they had changed things around in the code. It had some issues but I noticed 1.1.4.5 popped up on 3/1. I had some shows on encrypted channels that I wanted to record so I swapped back to windows and loaded up 1.1.4.5 and it's been up for 4 Days 22 Hours 2 Minutes 21 Seconds. Memory utilization is 12376 KB / 30256 KB.

I figure I'll let it run until it either crashes or hits 30 days uptime before giving 1.1.5.2 a try.

I've never been able to stay up for more than than 3 days 11 hours or so with 1.0.3.2 without a segfault or 2 days 5 hours or so without a memory leak or crash with 1.0.9.5 and 1.1.2.4.

I'm keeping my fingers crossed that the problem has been identified and fixed.

adam1991

Posts: 2893
Joined: Sat Jun 11, 2011 2:31 pm
Location:

HTPC Specs: Show details

#45

Post by adam1991 » Wed Mar 07, 2012 3:56 am

Good to know. Thanks.

Mine isn't broken (yet, anyway), so I'll just watch these threads for details. If there are other compelling reasons to move on this, I'd be interested to know.

My usual timeframe for mucking with things like this is June-August and the Christmas break.

User avatar
WhatHappend

Posts: 69
Joined: Mon Jun 06, 2011 8:00 pm
Location:

HTPC Specs: Show details

#46

Post by WhatHappend » Wed Mar 07, 2012 4:51 am

KHTPC75 wrote: It had some issues but I noticed 1.1.4.5 popped up on 3/1.
How did it show up to you on 3/1? In your ticket or email?
KHTPC75 wrote: I figure I'll let it run until it either crashes or hits 30 days uptime before giving 1.1.5.2 a try.
I would recommend going to the latest version ASAP. If you get the latest to crash in the next few days, they can incorporate a fix ASAP. If the 1.1.4.5 crashes, they will just have you try 1.1.5.2 before looking into it (I would do the same for my customers).
KHTPC75 wrote: I'm keeping my fingers crossed that the problem has been identified and fixed.
I ran Digital Cable Diagnostics Tool (DCDT) http://experts.windows.com/windows_medi ... 51891.aspx, and was able to do 4000 channel changes (1000 per tuner) with the all tuner scan with no lockups. These recent FWs are the first FWs since I have owned the InfiniTV4 (Nov 2010) that have not crashed during that test. I think this is a good sign that the root issue has been found.

KHTPC75

Posts: 47
Joined: Sat Feb 25, 2012 12:56 am
Location:

HTPC Specs: Show details

#47

Post by KHTPC75 » Wed Mar 07, 2012 5:05 am

WhatHappend wrote:
KHTPC75 wrote: It had some issues but I noticed 1.1.4.5 popped up on 3/1.
How did it show up to you on 3/1? In your ticket or email?
It's on their linux driver page at the bottom. I had problems compiling the code under 64 bit Linux so when I went to download the drivers again under 32 bit Linux, I noticed they had beta firmware 1.1.4.5 available.
WhatHappend wrote:
KHTPC75 wrote: I figure I'll let it run until it either crashes or hits 30 days uptime before giving 1.1.5.2 a try.
I would recommend going to the latest version ASAP. If you get the latest to crash in the next few days, they can incorporate a fix ASAP. If the 1.1.4.5 crashes, they will just have you try 1.1.5.2 before looking into it (I would do the same for my customers).
Good point. I may load it up tonight.
WhatHappend wrote:
KHTPC75 wrote: I'm keeping my fingers crossed that the problem has been identified and fixed.
I ran Digital Cable Diagnostics Tool (DCDT) http://experts.windows.com/windows_medi ... 51891.aspx, and was able to do 4000 channel changes (1000 per tuner) with the all tuner scan with no lockups. These recent FWs are the first FWs since I have owned the InfiniTV4 (Nov 2010) that have not crashed during that test. I think this is a good sign that the root issue has been found.
Very good news!

KHTPC75

Posts: 47
Joined: Sat Feb 25, 2012 12:56 am
Location:

HTPC Specs: Show details

#48

Post by KHTPC75 » Fri Mar 09, 2012 11:18 pm

KHTPC75 wrote:
WhatHappend wrote:
KHTPC75 wrote: I figure I'll let it run until it either crashes or hits 30 days uptime before giving 1.1.5.2 a try.
I would recommend going to the latest version ASAP. If you get the latest to crash in the next few days, they can incorporate a fix ASAP. If the 1.1.4.5 crashes, they will just have you try 1.1.5.2 before looking into it (I would do the same for my customers).
Good point. I may load it up tonight.
1.1.5.2 segfaulted on me after 2.63 days of uptime :( Diags sent to Ceton.

The crashlog.txt file has a lot of new debugging information so hopefully it's enough for them to locate and resolve the issue.

brainwarped

Posts: 10
Joined: Tue Feb 28, 2012 1:35 am
Location:

HTPC Specs: Show details

#49

Post by brainwarped » Wed Mar 21, 2012 9:52 am

I have been using firmware 1.1.6.9 since March 13th and I am at 6 days of uptime. It probably would have been longer, but windows update rebooted the 7mc machine.

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#50

Post by jwill42 » Thu Mar 22, 2012 9:44 pm

I just responded (on a long-open support ticket) to Ceton's request to let them know how my InfiniTV was doing with the updated FW (1.1.5.2) they had suggested I try a few weeks ago. Here is my response to Ceton:

***



I reverted my PCIe InfiniTV to old FW, HW, and drivers and have been successfully running now for 31 days without the firmware crashing. I have had 4 or more programs recorded every day, and no problems. Clearly the older FW and HW work fine (as I remembered, I'm not sure why you guys forced me to guess the URL to download the old FW that I needed), and Ceton caused a regression in later FW and HW versions that resulted in the crashing.

Here is what I have been running for over a month now:
firmware 1.0.2.2 2011-Feb-03
HW 4790
driver 1.0.1.21 2011-Jan-14

As I said, all of my recordings have worked perfectly, and I have 31 days of uptime on the PCIe InfiniTV. The only odd thing I noticed is that the Ceton diagnostic program says (and has said from the beginning of this trial):

"InfiniTV is not configured for use with Media Center. Please reinstall InfiniTV drivers"

Obviously that is incorrect, since the recordings have been working fine for a month in Media Center.

Now that I have found that my PCIe InfiniTV is stable with the February 2011 FW and HW, plus the January 2011 drivers, the next thing I need to do is get my USB InfiniTV working reliably (it has not been hooked up at all during the past month).

Questions for Ceton:

1) Am I correct in assuming that I will need to update to newer Ceton drivers in order for the USB InfiniTV to be recognized? If so, what version would you recommend? The latest Ceton drivers? Or something in between the latest and the January 2011 drivers I am running now?

2) After I update the Windows drivers, I am going to run the PCIe InfiniTV by itself for a while longer, to make certain it is still stable (I will keep the old FW and HW on the PCIe InfiniTV). Once I am satisfied that the PCIe InfiniTV is stable with the new drivers (and old FW), I will introduce the USB InfiniTV. What FW would you recommend I test on the USB InfiniTV? Obviously I cannot use the same Feb 2011 FW and HW that I am using on the PCIe InfiniTV. I guess I should probably test the latest beta FW that you have for the USB InfiniTV.

I see that there is a 20120312 1.1.6.9 FW image available on Ceton's website. Would you recommend I test that one? Or a newer one? What about the HW image, what should I use for HW when I test the USB InfiniTV with the new FW?

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#51

Post by jwill42 » Fri Mar 23, 2012 12:00 am

What hardware (HW) image (not firmware) do people have on their USB InfiniTV units? Have you received any links to beta HW images for USB InfiniTV?

I don't think the responses I got from Ceton are correct.

***

Ceton:
We would like you to try the 1.1.6.9 firmware (available at http://cetoncorp.com/firmware/ceton_inf ... _6_9.IMAGE), the 4884 hwimage and the latest drivers. Please let us know the outcome of this.
Me:
I was asking about my USB InfiniTV. I'm not crazy enough to change the FW and HW on my PCIe InfiniTV now that it is finally working again.

I don't think 4884 is the correct HW image for USB, is it? Please double check. And provide me a link to the HW image that I should use on the USB InfiniTV.
Ceton:
Sorry for the confusion - we thought you were talking about your PCI InfiniTV. On your USB InfiniTV, you should use the 4881 hwimage which is already on it.

erkotz

Posts: 1378
Joined: Mon Aug 22, 2011 9:23 pm
Location:

HTPC Specs: Show details

#52

Post by erkotz » Fri Mar 23, 2012 12:12 am

jwill42 wrote:What hardware (HW) image (not firmware) do people have on their USB InfiniTV units? Have you received any links to beta HW images for USB InfiniTV?
They are correct - I wrote them the response that you quoted. We have never shipped (either on units, or as a downloadable file) a hwimage other than 4881 for the USB InfiniTV.
Quality Assurance Manager, Ceton Corporation

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#53

Post by jwill42 » Fri Mar 23, 2012 12:29 am

erkotz wrote: They are correct - I wrote them the response that you quoted. We have never shipped (either on units, or as a downloadable file) a hwimage other than 4881 for the USB InfiniTV.
What about beta HW images? I seem to remember seeing someone posting that their USB InfiniTV was running a hardware image in the 5000's.

erkotz

Posts: 1378
Joined: Mon Aug 22, 2011 9:23 pm
Location:

HTPC Specs: Show details

#54

Post by erkotz » Fri Mar 23, 2012 12:36 am

jwill42 wrote:
erkotz wrote: They are correct - I wrote them the response that you quited. We have never shipped (either on units, or as a downloadable file) a hwimage other than 4881 for the USB InfiniTV.
What about beta HW images? I seem to remember seeing someone posting that their USB InfiniTV was running a hardware image in the 5000's.
There has never been a public beta image for the USB InfiniTV other than 4881. Additionally, if you open a support ticket with Ceton Support, please install the software/firmware versions that are requested. We often have reasons for requesting that you install specific things.
Quality Assurance Manager, Ceton Corporation

joevulture

Posts: 22
Joined: Sun Jan 01, 2012 3:48 am
Location: Indianapolis, IN

HTPC Specs: Show details

#55

Post by joevulture » Fri Mar 23, 2012 12:39 am

jwill42 wrote:I reverted my PCIe InfiniTV to old FW, HW, and drivers and have been successfully running now for 31 days without the firmware crashing. I have had 4 or more programs recorded every day, and no problems. Clearly the older FW and HW work fine (as I remembered, I'm not sure why you guys forced me to guess the URL to download the old FW that I needed), and Ceton caused a regression in later FW and HW versions that resulted in the crashing.

Here is what I have been running for over a month now:
firmware 1.0.2.2 2011-Feb-03
HW 4790
driver 1.0.1.21 2011-Jan-14

As I said, all of my recordings have worked perfectly, and I have 31 days of uptime on the PCIe InfiniTV. The only odd thing I noticed is that the Ceton diagnostic program says (and has said from the beginning of this trial):
Do you have issues with random pixelation in copy-once recordings as posted by a bunch of others (including myself) with that firmware, HW and driver?

-- Joe

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#56

Post by jwill42 » Fri Mar 23, 2012 1:17 am

erkotz wrote:Additionally, if you open a support ticket with Ceton Support, please install the software/firmware versions that are requested. We often have reasons for requesting that you install specific things.

After Ceton failed to help for months, I actually fixed the crash / reboot issue on my PCIe InfiniTV myself, by asking Ceton for an old FW / HW image from February 2011 (since I knew my PCIe InfiniTV was working well at that time), but you claimed no FW existed from that time. So I had to waste my time guessing the right URL from Ceton's website to download FW and HW images from February 2011 -- which do indeed exist -- then I flashed them onto my PCIe InfiniTV and ran a careful trial for 31+ days, finding that all of the crash/reboot problems went away. All of my recordings were complete, no partials, and no recording conflicts.

Then I kindly informed Ceton about my findings, so Ceton could track down the regression that Ceton introduced into their FW / HW sometime after the version that I just carefully tested. All Ceton has to do now is go through their changelogs to identify what they changed that introduced the crash/reboot regression.

Ceton does have changelogs and a revision control system on their firmware source code? Hopefully?

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#57

Post by jwill42 » Fri Mar 23, 2012 1:23 am

joevulture wrote:
Do you have issues with random pixelation in copy-once recordings as posted by a bunch of others (including myself) with that firmware, HW and driver?
I wasn't specifically watching for it, so I'm not sure. With the newer FW on my PCIe InfiniTV, I did see the horizontal line pixelation issue occasionally, but I think it was very rare (or else I did not always notice it), so the fact that I did not notice it during the past month should not be taken as any evidence that the older FW did not have the problem. The old FW may very well still have the pixelation problem. I will try to watch for it when I get a chance. But my top priority now is to get my USB InfiniTV working reliably (and keep my PCIe InfiniTV working reliably as it now has been for the past month). Once those two are no longer missing recordings, then I will worry more about video noise.

jwill42

Posts: 23
Joined: Fri Feb 24, 2012 3:32 am
Location:

HTPC Specs: Show details

#58

Post by jwill42 » Fri Mar 23, 2012 1:35 am

Here are people who say they have hardware (HW) image version 5000 or 5124 on their InfiniTV, but I think they may have PCIe cards:

http://www.avsforum.com/avs-vb/showthre ... st21718111

http://www.avsforum.com/avs-vb/showthre ... st21717985

joevulture

Posts: 22
Joined: Sun Jan 01, 2012 3:48 am
Location: Indianapolis, IN

HTPC Specs: Show details

#59

Post by joevulture » Fri Mar 23, 2012 1:38 pm

jwill42 wrote:After Ceton failed to help for months, I actually fixed the crash / reboot issue on my PCIe InfiniTV myself, by asking Ceton for an old FW / HW image from February 2011 (since I knew my PCIe InfiniTV was working well at that time), but you claimed no FW existed from that time. So I had to waste my time guessing the right URL from Ceton's website to download FW and HW images from February 2011 -- which do indeed exist -- then I flashed them onto my PCIe InfiniTV and ran a careful trial for 31+ days, finding that all of the crash/reboot problems went away. All of my recordings were complete, no partials, and no recording conflicts.

Then I kindly informed Ceton about my findings, so Ceton could track down the regression that Ceton introduced into their FW / HW sometime after the version that I just carefully tested. All Ceton has to do now is go through their changelogs to identify what they changed that introduced the crash/reboot regression.

Ceton does have changelogs and a revision control system on their firmware source code? Hopefully?
Wow... Just wow.

I personally appreciate the effort that you have gone to in determining what happened when, even if nothing comes of it. (I know that neither erkotz, nor anybody else at Ceton can really publicly comment on this.) My day job is a software developer (embedded devices, actually), and I wish that I had a test team that had the resources to put in this much regression testing.

-- Joe

lurch66

Posts: 244
Joined: Fri Jun 10, 2011 4:32 am
Location: Queen Creek AZ

HTPC Specs: Show details

#60

Post by lurch66 » Sat Mar 24, 2012 5:00 am

has anybody tried firmware 1.1.7.2 and if so how is it working? Thanks in advance for any feedback I am just wondering before I install it, Thanks.

Post Reply