Crash (every 3-4 days) when "phoning home"

[jefbear]

Hi Guys,

I have had my Tivo for about a month and it works great BUT every couple of days it crashes and deletes ALL stored data. I am seriously going crazy.

It looks like the issue happens during a failed "phone home". What happens is:

1. Phone home fails
2. I log into tivoweb and see the phone call failed "unknown reason"
3. Start the call again & CRASH/REBBOT
4. It boots and works for 1 min (I think it tries the call again) CRASH/REBOOT
5. Does #4 3-4 times
6. Finally boots and goes into"deleting all program info..."
7. Comes back and whoo-hoo It works!
8. I manually start the phone home
9. I have to RELOAD my (backed-up) season passes and wishlists
10. The "backed-up" season passes channels are ALL mapped to HITS830 channel (this is a radio station!)
11. I remapp the channels to the right ones
12. I pray it stops doing this!!

Has anyone have any info they can give me to help me out here?
All info will be VERY much appreciated!!!


Cheers,

[tenty]

The only one thing I could suggest is looking though the errLog to see what is causing your problems.

Is this a stock standard install with the most recent OzTiVo image (1.3 at time of writing)? and happens right after a re-image of the drive? or have you loaded any new hacks?

If we can get some more information about your problem, we could better troubleshoot it!

Cheers
-Edwin

[jefbear]

Thanks for the quick response!
If you need anything else please be specific and I will do my best to get it


Ok, let me see if I can answer your points:

1. Here is the OTVERR log (TVerr was 0 bytes):
Dec 1 18:42:26 (none) EvtSwitcher[79]: MFS-using group 178 died!
Dec 1 18:43:18 (none) TmkTransform::Trace[142]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 18:43:18 (none) TmkTransform::Trace[142]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 18:43:18 (none) last message repeated 2 times
Dec 1 18:44:56 (none) EvtSwitcher[79]: MFS-using group 177 died!
Dec 1 18:45:50 (none) TmkTransform::Trace[142]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 18:45:50 (none) TmkTransform::Trace[142]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 18:45:50 (none) last message repeated 2 times
Dec 1 18:47:25 (none) EvtSwitcher[79]: MFS-using group 177 died!
Dec 1 18:48:18 (none) TmkTransform::Trace[142]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 18:48:18 (none) TmkTransform::Trace[142]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 18:48:18 (none) last message repeated 2 times
Dec 1 18:50:01 (none) EvtSwitcher[79]: MFS-using group 177 died!
Dec 1 18:50:54 (none) TmkTransform::Trace[142]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 18:50:54 (none) TmkTransform::Trace[142]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 18:50:54 (none) last message repeated 2 times
Dec 1 18:52:29 (none) EvtSwitcher[79]: MFS-using group 177 died!
Dec 1 18:53:23 (none) TmkTransform::Trace[142]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 18:53:23 (none) TmkTransform::Trace[142]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 18:53:23 (none) last message repeated 2 times
Dec 1 18:54:59 (none) EvtSwitcher[79]: MFS-using group 177 died!
Dec 1 19:15:11 (none) TmkTransform::Trace[143]: Have a buffer (0) with zero segmentsDid someone zero memory?
Dec 1 19:15:11 (none) TmkTransform::Trace[143]: leaking buffers = pt->theBuf = 0x30301000. Trying to recover.
Dec 1 19:15:11 (none) last message repeated 2 times
Dec 1 19:15:11 (none) last message repeated 2 times

2. This was imaged from the latest image (March 2004) from here:
http://minnie.tuhs.org/TiVo/files/images/iso/

3. Yes I have a couple hacks (but not too many).
I'm not sure how to see what exactly I have added (is there an easy way to tell?), but here are the ones I can remember:
1. Backup (latest version)
2. Channel grid
3. Hack manager (useless really)
LIST:
Main Menu, Backup, Channel Grid, Conflict Resolve, Grid, Hack Manager, Info, Logos, Logs, Mail, MFS, MfsStream, PAL NTSC, Phone, Reboot TiVo, Remount Root, Resource Editor, Schedule, Screen, Search, Space Used, Theme, User Interface, Web Remote, What's On, Restart


Thnaks Edwin!!!


CHeers,

[jefbear]

Here are a couple more debugs/logs.

It looks like the last one was the Tivo dumping the "backed-up" seson passes
I had to restore them again!


Thanks heaps!


Cheers,

[DavidSymons]

Hi Jeff,

I couldn't see much wrong with these files (others may differ).

Could you post more of your /var/log/tverr? The "Trying to recover" errors you've already posted from it look bad but they're present on most normally functioning OzTiVos. There might be something else in this file.

Also the kernel log (/var/log/kernel) might be worth a look too.

Cheers, Dave.

[jefbear]

Hi Dave,


1. Unfortunatly, (I think) most of these logs are wiped when the crash happens. I assume this because most of the logs are only one day long and very small.

As I noted (and may have been missed the tverr was/is ZERO K or has no entires!


2. Here is the kernal (looks like it was wiped too):
Dec 2 18:00:28 (none) kernel: CodeFromNec(4fb08f0) is meaningless
Dec 2 18:00:32 (none) kernel: CodeFromNec(a55a1ce0) is meaningless
Dec 2 18:00:32 (none) last message repeated 2 times
Dec 2 18:00:47 (none) kernel: CodeFromNec(4fb0bf0) is meaningless
Dec 2 18:01:02 (none) last message repeated 4 times
Dec 2 18:01:10 (none) kernel: CodeFromNec(a55a0ff0) is meaningless
Dec 2 18:01:10 (none) last message repeated 2 times
Dec 2 18:01:19 (none) kernel: CodeFromNec(4fb0bf0) is meaningless
Dec 2 18:01:22 (none) kernel: CodeFromNec(4fb0bf0) is meaningless
Dec 2 18:01:39 (none) kernel: CodeFromNec(a55a0bf0) is meaningless
Dec 2 18:01:40 (none) last message repeated 3 times
Dec 2 18:53:28 (none) kernel: CodeFromNec(4fb0bf0) is meaningless

Here are the logs:
cron.log----8553----Dec 02 13:04:28
endpad.log---3394---Dec 02 12:44:54
kdebug----0----Dec 02 13:24:30
kernel ----759 ---- Dec 02 18:53:28
maillog ---- 0 ---- Jan 01 00:00:37
messages ----47 ---- Dec 02 13:24:30
secure ---- 0 ---- Jan 01 00:00:37
svclog ---- 747 ---- Dec 02 13:24:38
tcdebuglog ----69 ---- Dec 02 13:24:39
tclient ---- 1304 ---- Dec 02 13:24:48
tivoLog.prv ----2275 ---- Dec 02 19:00:31
tivoLog.prv.gz ----1392 ----Dec 02 13:04:54
tivoLog.pub ----0 ---- Dec 02 13:04:52
tvdebuglog ----0 ---- Dec 02 13:24:30
tverr ---- 0 ---- Dec 02 13:24:30
tvlog ---- 64466 ----Dec 02 18:59:32

Oendpad.log ----4183 ---- Dec 02 12:33:28
Okdebug ---- 0 ---- Dec 01 19:53:18
Okernel ---- 100027 ----Dec 02 13:24:08
Omessages ----25727 ---- Dec 02 13:06:21
Otcdebuglog ----33130 ---- Dec 02 13:24:30
Otclient ---- 62863 ----Dec 02 13:24:30
Otvdebuglog ----0 ---- Dec 01 19:53:19
Otverr ---- 3488 ---- Dec 02 13:24:30
Otvlog ---- 481002 ----Dec 02 13:24:30

I actually imaged 2 100Gig drives at the same time a month ago. This one is the second drive because I totally messed up the first one.

Maybe another reimage is in order? (I really hate this thought!)


Thanks all!!!


Cheers,