Can't get CacheCard program data on"successful"daily call; OK on full "clear & delete

[digbydel]

I have to run a full "clear and delete everything" every two weeks on my Tivo 1 before I get up to date Programme Info - it was working OK until my pal fitted the CacheCard and we connected thru the web. Do you experts have any ideas please re what we've done wrong? And can you tell us: does the programme data come from different sources (ip address) according to the UK geo/region?

Until the CacheCard was installed, my telephone dialup service and programme update worked fine. But I do need TivoWeb access and I wanted to remove the dependence on the telephone line. My UK Tivo 1 has two discs: 40 and 120mb; my pal is a "relatively new" Tivo "expert" ;o) - he's a lovely bloke and he's very upset that I'm now having problems - he installed my CacheCard and made the necessary software/code changes by removing the h/disk and placing it in a PC chassis.

He made the necessary changes at his home (220 miles from my West London home - in Bury near Manchester); we then connected to his network and we got new programme data - 4 weeks of programme data - great!! - much more than I ever got via my telephone update - and we had no problems. When I got the Tivo back to my West London home, I connected to the network, and was happy that it was (apparently) making "successful" daily calls; but I hadn't noticed that I was "not getting any new programme updates" (even though the daily call was described ech day as being "successful").

I eventually ran out of programme info.. and I've found that the only way I can get the programme data refreshed, is by doing a full "clear and delete eveything" within "System Reset". Any action less than that will NOT refresh my programme data. Needless to say this is a drastic action for a Tivo; it means that I have to reinstall and retrain it 2/3 weeks re all our favourites and preferences.

NB: My colleague in Bury nr Manchester always gets more weeks' programme data than I do, even when I do a full "clear and delete everything". He will typically have a month or maybe slightly more of programmes; whereas I will tend to get 3 weeks' data. (This might make sense to you; it doesn't to me. But it makes me wonder whether my daily call is targetting an incorrect IP address.)

(I'm using a Pace On-digital receiver - and receive Freeview only - I select Basic A as my programme group. My Tivo previously worked well on this setup via telephone dialup; but now that the CacheCard is fitted I find that I cannot now revert to telephone updates ie. it just won't connect. I don't want to revert to telephone updates, TivoWeb is clearly the way to go.)

Does anyone have a magic wand please?

Thanks for any advice/direction you can give me.

[blindlemon]

Firstly, are you sure it's actually connecting over the web?

Go in via telnet and type

cat /etc/rc.d/rc.net

and past the result in your reply here.

Secondly, you may have a swapfile (or lack of) problem. You can check this from TiVoWeb on the Info menu - just see what it says for Swap.

[ericd121]

blindlemon, can you remind us (me), what the command is to set your IP Address, Subnet Mask, and MAC Address?

[mike0151]

For Turbonet cards (and presumably for cachecards) it is, at the bash prompt :-

nic_config_tivo

This is usually in /sbin and will normally be in the path.

HTH

Mike

[Glesgie]

I seem to remember that this can happen if the mtu on your router is set to less than 1500.

Apparently the TiVo's initial setup call is different to the daily call.

Hope this helps.

[digbydel]

Dear all

I'm really grateful to you all for your initial thoughts and suggestions,...
here's a data dump which I hope addresses them,..

1. First we changed MTU to 1500 - that's the maximum per this router.
MTU had been 1458; changed to 1500 and rebooted router.

We then we ran another Daily Call,... we ended with a "Pending Restart" (a Restart is always required whenever we trigger a "Daily Call" - is this normal?

Result: no change to programme data. eg. BBC 1, & 2 ITV 1 still no data; Channel 4?progs up to Feb 17 then no more data. Successfull call says programme data up to Feb 24 (not sure on which channel it actually holds Feb 24 data, but,...). (I know that a full "clear and delete everything" will give me all the missing data, but I don't want to that right now as I want to stay in a condition when I can see when it has been fixed.
_______________

Swap shows 130gb with 99+% is free

_________
IPCop (firewall) is showing the Tivo talking to:

192,168,121,241:1028 --> 204,176,49,3:80
192,168,121,241:1027 --> 204,176,49,3:80
192,168,121,241:1026 --> 204,176,49,3:80
192,168,121,241:123 --> 204,176,49,12:123
192,168,121,241:123 --> 204,176,49,11:123
192,168,121,241:123 --> 204,176,49,10:123

_______________________

Thanks in advance for any more thoughts / ideas.
Kind regards, Digbydel

[digbydel]

oh, and the cat command shows this:

#
#!/bin/sh
#

if /sbin/insmod -f /lib/modules/turbonet2.o macaddr=00: 0B: AD: 14: 68: 61 ; then
export DYNAMIC_NET_DEV=eth0
/sbin/ ifconfig eth0 (my ip address) netmask (my netmask) up
/sbin/ route.tivo add default gw (my ip address) netmask (my netmask)fi
bash-2.02#
bash-2.02#

[blindlemon]

If you are connecting to the internet via a router then the gateway needs to be set to your router's IP address, not that of your PC.

[digbydel]

Ok. Sorry, the gateway ip *is* set to that of the router (192.168.121.254)

if /sbin/insmod -f /lib/modules/turbonet2.o macaddr=00:0B:AD:14:68:61 ; then
export DYNAMIC_NET_DEV=eth0
/sbin/ifconfig eth0 192.168.121.241 netmask 255.255.255.0 up
/sbin/route.tivo add default gw 192.168.121.254 netmask 0.0.0.0
fi

The Tivo does seem to connect somewhere - to 204.176.49.x range on port 80 and 123.

[digbydel]

Seem to be errors in this log file from the nic_config_tivo command. :confused:

[blindlemon]

I can't see anything from nic_config_tivo but it looks like you have requested the 2.5.5a download - maybe with an LBA48 kernel installed?

Try typing

cat /proc/version

to see what the kernel version is. If you have the standard kernel it will say Linux version 2.1.24-TiVo-2.5 (build@buildmaster17) (gcc version 2.8.1) #2 Thu Feb 7 17:49:48 PST 2002whereas the LBA48 kernel will show up as Linux version 2.1.24-TiVo-2.5 (millert@xerxes.courtesan.com) (gcc version 2.8.1) #14 Wed Oct 8 12:06:25 MDT 2003You don't need the LBA48 kernel with the drives you have, but if you do need 2.5.5a then I would suggest doing a 'self-upgrade' as described in this thread (http://www.tivocommunity.com/tivo-vb/showthread.php?t=248502) with the files from here (http://www.tivoheaven.com/download/255a.zip). That should stop the TiVo OS from trying to apply the download whenever it calls home and you may find that your daily calls start working again.

[digbydel]

OK. I have the standard kernel. ;)

I went to download the 255a.zip from TivoHeaven but it's corrupt . . . any ideas?

THanks for all your help

[Ian_m]

OK. I have the standard kernel. ;)

I went to download the 255a.zip from TivoHeaven but it's corrupt . . . any ideas?

THanks for all your help
Use IE to download ?

[blindlemon]

I went to download the 255a.zip from TivoHeaven but it's corrupt . . . any ideas?It's fixed now - for some reason the file seemed fine and would download OK with IE and FTP but was truncated when downloaded with Firefox.

I re-uploaded it manually and it now works fine... :confused:

[digbydel]

OK. Still in the same position as yesterday. When you make the daily update call it says it's been 'sucessful' but always comes back as 'pending restart'. When you do the restart it says it's installing new software from Tivo. It never actually gets the program data.

I followed blindlemon's advice to the letter and copied up the pxmpegdecode.o file, even to the point of copying the build-version file too. Done it several times too - before and after the daily call / pending restart messages. Tivo says that it's 2.5.5 rather than 2.5.5a on each occasion also.

I've also switched it back to dialup for the calls. That doesn't help either.

I've attached a couple of the log files (after a reboot and a daily call), but whats going through my mind at the moment is a complete wipe & reinstall of the Tivo disks <gulp>.

As aways thanks for all your help guys.

[digbydel]

I just wanted to be sure that you guys realise that I haven't ever "requested to upgrade to 2.5.5a-01-1-023" ie. the action recorded on my Tivo_log which shows "that my Tivo 1 was suddenly trying to install 2.5.5a-01-1-023" was triggered by persons/systems unknown :) - presumably by the Tivo service.

Seems strange that this has never happened before ie. first attempt was yesterday Feb 16 at 10:00am; noone was in the house at the time.
The only other action listed on this on my Tivo-log
(note that the only other action documented on my Tivo_log happened on Jan 1.

Thanks for any further advice/direction

[blindlemon]

For some reason your TiVo seems to have downloaded 2.5.5a even though it is supposed to be on 2.5.5. Your pal didn't re-image your drives from his (2.5.5a version) backup did he?

Anyway, as it already thinks you have 2.5.5a, it might be worth editing the build-version file with joe to remove the "a" before trying another call. Having the "a" in the file if you are supposed to be on 2.5.5a but don't want a download seems to work, so maybe the opposite will work too...

[digbydel]

OOOOkay . . .

Editing the build-version file did nothing to help. In the Kernel Log it shows this:

Feb 18 17:35:36 (none) kernel: "foreach package $packages {
Feb 18 17:35:36 (none) kernel: puts "Install $package into $installdir"
Feb 18 17:35:36 (none) kernel: exec jpm -p $package -i $installdir
Feb 18 17:35:36 (none) kernel: }"
Feb 18 17:35:36 (none) kernel: (file "/var/utils/updateroot" line 114)
Feb 18 17:35:36 (none) kernel: child process exited abnormally
Feb 18 17:35:36 (none) kernel: while executing
Feb 18 17:35:36 (none) kernel: "exec /var/utils/updateroot /dev/hda /install /var/packages >&@ stdout"
Feb 18 17:35:36 (none) kernel: (object "::inst" method "::Installer::installSoftware" body line 46)
Feb 18 17:35:36 (none) kernel: invoked from within
Feb 18 17:35:36 (none) kernel: "inst installSoftware $name $Inc::TC_INSTALL_DIR "
Feb 18 17:35:36 (none) kernel: (file "/tvbin/installSw.itcl" line 182)
Feb 18 17:35:36 (none) kernel: Could not check software installation: child process exited abnormally
Feb 18 17:35:36 (none) kernel: Scanning for phase3 repair scripts
Feb 18 17:35:36 (none) kernel: Starting Services.
Feb 18 17:35:36 (none) kernel: Scanning for phase4 repair scripts
Feb 18 17:35:37 (none) kernel: MCP startup complete
Feb 18 17:35:37 (none) kernel: rc.sysinit is complete

Its seeing that 2.5.5a is there but not active.

Any ideas?

I know it's not in the spirit of things, but I'm totally lost. Is there anyone I can pay to dial in and have a look?

D

[blindlemon]

Well, as much as I enjoy a challenge, I suspect this might be a good time to revert to plan B - ie. re-image the drive from scratch from a known-good backup.

As you have been doing a C&DE every couple of weeks anyway, there can't be much to lose at this point.

[Glesgie]

Just a thought, but have you checked to see if there is anything in /var/packages/ ?

If there is then it could be a "bad" software update and deleting might help.

[digbydel]

Tried deleting the files in /var/packages.

Where do I get a 'good known backup' from?!

Thanks!

[steveroe]

Where do I get a 'good known backup' from?!

Thanks!

Post your request in this (http://www.tivocommunity.com/tivo-vb/showthread.php?p=3789960) thread