6.3f is here

[w2kr]

I noticed my Series 2 DirectTv TiVos now have version 6.3f ready to install. Two versions show up in my MFS:

6.3f-01-2-101 tyDb 3395068 12/08/07 03:11 780
6.3f-01-2-121 tyDb 3395073 12/08/07 03:11 780

Perhaps this is the fix for the 6.3e problems?

[ronnythunder]

strange - what model do you have? not on any of my boxes.

ronny

[w2kr]

It's an RCA DVR40

[crashHD]

I've got the same thing on my one of my DVR40's, nothing on the other, and nothing on the R10's yet

[labbie48]

Just checked all of mine and 6.3f is not on any of them. Will check every few days....

[RandC]

6.3f in MFS on HR10-250 12/07/07

[crashHD]

Perhaps this is the fix for the 6.3e problems?

The fix for 6.3e is 6.2a, unless you're running an Hr10, in which case, 6.3f is hopefully the fix.

[RandC]

files removed or changed from 6.3e to 6.3f

Changed
tivoapp

Removed
dhclient /sbin
rtl8150 /lib/modules
kaweth /lib/modules
dhclient-script /etc

[RandC]

POSSIBLE Confirmed new location patches 6.3f only. This information is for persons with higher pay grades to confirm.
DO NOT try these locations without confirmation from the higher authorities as I used a very crude method to locate the possible new locations. Unless of course you want to screw your unit.

#No encryption (DEC) 1602336 orig 0c16a5ed
#30-second skip (DEC) 6707932 orig 1440002b
#Backdoors (DEC) 2804772 orig 00008021
#HMO/HME
(DEC) 903776 orig 00408821
(DEC) 1118652 orig 00408021
(DEC) 5705052 orig 14400014
#Free Trial (DEC) 918052 orig 0c144461
#OSD Mesage (DEC) 6719940 orig 0c319c9e

[100Tbps]

I noticed my Series 2 DirectTv TiVos now have version 6.3f ready to install. Two versions show up in my MFS:

6.3f-01-2-101 tyDb 3395068 12/08/07 03:11 780
6.3f-01-2-121 tyDb 3395073 12/08/07 03:11 780

Perhaps this is the fix for the 6.3e problems?

I'm on 6.3e and even without a reboot in 30+ days, I haven't noticed any problems. What have you experienced?

[whitepelican]

I think I've had about one reboot in the past 6 months on my HR10-250 that wasn't caused by ffmpeg created ty files. As far as I'm concerned 6.3e on the HR10-250 is now about as stable as 6.2/6.2a were on the SD DirecTivos.

[wrasp]

fsid type name
-----------------------------------
3095912 tyDb 6.3c-01-2-357
4349308 tyDb 6.3f-01-2-357
3095912 tyDb ACTIVE

[Gunnyman]

can anyone verify whether the Tivo Branded wireless G adapter still works with 6.3f?

[RandC]

can anyone verify whether the Tivo Branded wireless G adapter still works with 6.3f?From ubs.map file
device usb-cdc
product 0a5c bd11 CDC Downloader
product 0a5c 0cdc CDC Proto board
product 150a 1010 TiVo Wireless G
product 150a 1011 TiVo Wireless G

[Gunnyman]

Thanks much :)

[crashHD]

6.3f slicepack, includes all SD series2, HR10, and R10.

6.3f (ed2k://|file|6.3f(6.3f-01.2_DirecTV_Tivo_slices).tgz|29018999|02EBAB75807C4D5367A037885862D0C0|h=SNW6LXM4ZWHZAVGGZBJLDH3P3A JJT26O|/)

[labbie48]

6.3f slicepack on rapidshare. read .nfo file included in tgz......

http://rapidshare.com/files/76862446/6.3f_6.3f-01.2_DirecTV_Tivo_slices_.tgz

[dt1401]

POSSIBLE new location patches 6.3f only. This information is for persons with higher pay grades to confirm.
DO NOT try these locations without confirmation from the higher authorities as I used a very crude method to locate the possible new locations. Unless of course you want to screw your unit.

#No encryption (DEC) 1602336 orig 0c16a5ed
#30-second skip (DEC) 6707932 orig 1440002b
#Backdoors (DEC) 2804772 orig 00008021
#HMO/HME
(DEC) 903776 orig 00408821
(DEC) 1118652 orig 00408021
(DEC) 5705052 orig 14400014
#Free Trial (DEC) 918052 orig 0c144461
#OSD Mesage (DEC) 6719940 orig 0c319c9e

Yes those all work. there now posted on dvrpedia.

http://www.dvrpedia.com/Modify_TiVoApp_to_disable_encryption_and_add_other_features

Disable encryption:

echo -ne "\x3C\x02\x00\x00" | dd conv=notrunc of=tivoapp bs=1 seek=1602336
30sec skip:

echo -ne "\x10\x40\x00\x2b" | dd conv=notrunc of=tivoapp bs=1 seek=6707932
Backdoors

echo -ne "\x24\x10\x00\x01" | dd conv=notrunc of=tivoapp bs=1 seek=2804772
HMO/HME

echo -ne "\x34\x11\x00\x00" | dd conv=notrunc of=tivoapp bs=1 seek=903776
echo -ne "\x24\x10\x00\x01" | dd conv=notrunc of=tivoapp bs=1 seek=1118652
echo -ne "\x10\x00\x00\x14" | dd conv=notrunc of=tivoapp bs=1 seek=5705052
removes *TiVo Plus features (trial ends today)

echo -ne "\x24\x02\x00\x00" | dd conv=notrunc of=tivoapp bs=1 seek=918052

[Callindril]

Assuming that MRV wasnt put back in.... :(

[100Tbps]

Has anyone determined what problems or bugs 6.3f resolves?

[crashHD]

Bufferhack works for 6.3f using the same data as for 6.3eset sys(6.3f) [list 0x0d6af2 0x0d690a 0x71 0x19e7e6 22029540 A9846DD76AEF86CEA6B3E69CE3758227598991A0]
This was tested on 6.3f-01-2-521 on an R10. I believe other units, including the HR10, will be the same, but I cannot confirm this.

[Rorschach]

Can someone hassle making a diff file so the dumbasses like me can apply it with superpatch?

[crashHD]

I just made this for my box#!/bin/sh

echo "Patching 6.3f Tivoapp"
echo "Copying /tvbin/tivoapp /tvbin/tivoapp.tmp"
cp /tvbin/tivoapp /tvbin/tivoapp.tmp
echo "Done copying, now patching tivoapp.tmp"

# Disable Encraption:
echo -ne "\x3c\x02\x00\x00" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=1602336
# Enable 30sec skip:
echo -ne "\x10\x40\x00\x2b" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=6707932
# Backdoors:
echo -ne "\x24\x10\x00\x01" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=2804772
# HMO/HME:
echo -ne "\x34\x11\x00\x00" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=903776
echo -ne "\x24\x10\x00\x01" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=1118652
echo -ne "\x10\x00\x00\x14" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=5705052
# Remove "*Tivo Plus features (trial ends today)"
echo -ne "\x24\x02\x00\x00" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=918052
# Enable on-screen clock (SPS9S toggle)
echo -ne "\x10\x40\x00\x1e" | dd conv=notrunc of=/tvbin/tivoapp.tmp bs=1 seek=176752


echo "Moving the running /tvbin/tivoapp to /tvbin/tivoapp.orig.bak"
mv /tvbin/tivoapp /tvbin/tivoapp.orig.bak
echo "Done that, moving /tvbin/tivoapp.tmp /tvbin/tivoapp"
mv /tvbin/tivoapp.tmp /tvbin/tivoapp

exit


cat that into an .sh file, and execute it. Note that it has none of the safety checks of superpatch. It is for 6.3f, if you run it on anything else, it will rape your tivoapp, and you'll be pulling the drive.
Also, you need to manually mount root rw for this to run. It doesn't do it for itself.
This also includes the code to activate the on-screen clock in the lower right hand corner of the screen. If you don't want that, comment it out or delete it prior to running the script.

[crashHD]

Here is a superpatch for 6.3f. Satisfaction guaranteed, or your money back!!

(Seriously, it's been lightly tested at this point. Use at your own box's peril!)

[RandC]

Bufferhack works for 6.3f using the same data as for 6.3eset sys(6.3f) [list 0x0d6af2 0x0d690a 0x71 0x19e7e6 22029540 A9846DD76AEF86CEA6B3E69CE3758227598991A0]
This was tested on 6.3f-01-2-521 on an R10. I believe other units, including the HR10, will be the same, but I cannot confirm this.confirming HR10-250 uses the same data.

[stevel]

My HR10 still does not have 6.3f in MFS. Curious.

[crashHD]

confirming HR10-250 uses the same data.
Thanks.

Are you referring to the superpatch and bufferhack, or just bufferhack?

My HR10 still does not have 6.3f in MFS. Curious.
Only one of my series 2 units has anything. DirecTV's deployment strategy is curious. I can see not hitting(install) all units everywhere, simultaneously, but why not download them all the first time?

[RandC]

Thanks.

Are you referring to the superpatch and bufferhack, or just bufferhack?Bufferhack only, I like to WinHex my tivoapp then upload.

My HR10 still does not have 6.3f in MFS. Curious.Same thing here, one of my units received slices, the other two did not. I downloaded slice pack from link a few posts back and dbload into my testing unit and installed there. All is good so far, no problems. It does appear that there is plenty of additional logging in the tvlog.

[tvtyme]

Here is a superpatch for 6.3f. Satisfaction guaranteed, or your money back!!

(Seriously, it's been lightly tested at this point. Use at your own box's peril!)

Works like a charm. Thanks, crashHD!

[jsauser11]

Howdy,
I have a hacked HR10-250 currently on 6.3d (just copied 6.3d tivoapp to a 6.3c install). I have been trying to get the 6.3f slices to download, but after forcing multiple phone calls, they do not appear is SWSystem. Here is what my SWSystem looks like:

Directory listing of /SwSystem
Name Type Id Date Time Size
6.3b-01-2-357 tyDb 2699182 12/13/06 13:55 796
6.3e-01-2-357 tyDb 3729185 08/11/07 02:36 780
ACTIVE tyDb 2699182 12/13/06 13:55 796

I have manually downloaded the 6.3f slices, but I can't seem to find instructions on how to use dbload to get them into the SWSystem on my unit so I can begin the manual update. Can someone point me in the right direction to instructions for dbload?

Thanks,
Jeff:eek:

[RandC]

Can someone point me in the right direction to instructions for dbload?

Thanks,
Jeff:eek:telnet into your TiVo then enter
dbload "slicename" there is GZcore, GZhpk-Series2, GZkernel-Series2, utils and be sure to use swsystem-115422985-2.slice for the HR10-250. Can take some time to load files into MFS so be patient

[jsauser11]

telnet into your TiVo then enter
dbload "slicename" there is GZcore, GZhpk-Series2, GZkernel-Series2, utils and be sure to use swsystem-115422985-2.slice for the HR10-250. Can take some time to load files into MFS so be patient
Thanks. Does the 6.3f .tar file have to be extracted first so the slices are in their uncompressed state, and which directory do they go into (/var/packages?). Is dbload called from the directory the slices are located in? Sorry for all the questions.
Jeff

[RandC]

Thanks. Does the 6.3f .tar file have to be extracted first so the slices are in their uncompressed state, and which directory do they go into (/var/packages?). Is dbload called from the directory the slices are located in? Sorry for all the questions.
Jeffextract tar file to get the individual slice files.
I put the files in my /var directory.
go to the directory where you put files and dbload from there.

[crashHD]

The slices in that tarball are also gzipped. After extracting the tar, gunzip them.

gunzip slicefilename

It's only necessary to gunzip the ones you need to use.

[alreadylost]

Software System: 6.2-01-2-351

Hughes SD-DVR40, with a 160gb drive, I currently use TIVOWEBPLUS 2.0, and MRV regularly. Two questions:

Is there any rush to upgrade to 6.3f?
Can I just leave my TIVO as is?

Thanks in advance for the help.

Jose'

[whitepelican]

Software System: 6.2-01-2-351

Hughes SD-DVR40, with a 160gb drive, I currently use TIVOWEBPLUS 2.0, and MRV regularly. Two questions:

Is there any rush to upgrade to 6.3f?
Can I just leave my TIVO as is?

Thanks in advance for the help.

Jose'

I wouldn't. Especially if you want to keep MRV, as it doesn't exist in the 6.3x line. What you should probably do is upgrade to 6.2a, though, so you will get the correct Daylight Savings Time changes and not have your Tivo's clock off by an hour several weeks of the year.

See this post at Tivocommunity.com (http://www.tivocommunity.com/tivo-vb/showthread.php?t=343660) for an easy way to upgrade to 6.2a and keep your box hacked as is.

[tvtyme]

Software System: 6.2-01-2-351

Hughes SD-DVR40, with a 160gb drive, I currently use TIVOWEBPLUS 2.0, and MRV regularly. Two questions:

Is there any rush to upgrade to 6.3f?
Can I just leave my TIVO as is?

Thanks in advance for the help.

Jose'

If you applied the 6.2 tivoapp patch to record/pause XM audio channels using the remote then you'll lose that, too. To date, no one's posted a 6.3x tivoapp patch for it.

[RandC]

I noticed today in svclog file the files for 6.3f were listed, on both of my units. One had them from the original download, the other I had dbloaded the files, both units are still 6.3e. My other unit I had to dbload the files and I installed 6.3f but the log show no entries for files, and did not for 6.3e either. Unless I am mistaken the odd thing is the file status was listed as success, I thought it should show failed because it was already in MFS.
aval_file TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505627 AVAL_ID=0000365200003846 FILE_NAME=BS-standard-p13903.slice.gz FILE_STATUS=SUCCESS SRC=
aval_file TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505640 AVAL_ID=0000365200003846 FILE_NAME=GZhpk-Series2-115422619-2.slice.gz FILE_STATUS=SUCCESS SRC=
aval_file TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505649 AVAL_ID=0000365200003846 FILE_NAME=GZkernel-Series2-115422594-2.slice.gz FILE_STATUS=SUCCESS SRC=
aval_file TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505653 AVAL_ID=0000365200003846 FILE_NAME=utils-115422371-2.slice.gz FILE_STATUS=SUCCESS SRC=
aval_file TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505653 AVAL_ID=0000365200003846 FILE_NAME=TiVo plany Data Download FILE_STATUS=FAILURE_0000FFFF SRC=
aval_pp TCD_ID=3570000B0AE8531 CALL_ID=1199931717 TIME=1201505654 AVAL_ID=0000365200003846 STATUS=SUCCESS SRC=plany

[wundernaut]

Why would DirecTV be pushing out a bug fix now when (supposedly) the major release with extra features we were promised last July for "early 2008" is just around the corner?

[crashHD]

bcuz them are the not make sense.

[tas3086]

6.3f has arrived in MFS. Unfortunately, the current version is very old. As a result, I've got a number of versions 6.2 6.3a,.... installed. 5 older versions in all. Can I just install the latest version, or do I have to install each of the older versions in sequence?

I'd believe that the code would be ok, but not sure that any and all MFS mods would be made if is skipped certain levels.

[crashHD]

Unless you manually loaded some slices, you don't have 6.2 and 6.3a on the same machine. That is a combination of slices that would not naturally exist in the wild.

You can go directly to 6.3f, if you want to.

Be sure that's what you want.

[Barefooter]

Has anyone else had issues with 6.3f After I upgraded an HR10 each time I hit Messages and Setup the machine reboots. Machine seems to be stable other than that but that is an issue.

[samhammer]

I have a hacked R10-250. I have not done anything with it for qhite a while. It still has 6.3a on it. But now I have people staying with me and they are using it in their room. It keeps rebooting so I wanted to upgrade it to 6.3f

I downloaded the slices and used dbload, but the slices still do not show up in the SWSystem directory ( as viewed from TivoWebPlus)

When I do the dbload, I do not get any errors. It takes a short time for each file and then I get the prompt again.

Any ideas on why the slices do not seem to load as planned?

Thanks
Sam

[crashHD]

Most likely, you missed one, or got them out of order.

you need to dbload:

GZcore
GZhpk-Series2
GZkernel-Series2
utils

Last you need to dbload the correct swsystem-xxxxxxxxxxxx file

The other 4 files must be dbloaded on the system (in any order)before swsystem. It must be last. Also, any slices in which there are 2 of them, one with -Series2- in the slice name, and another with -Gen04- in the slice name, you will need to dbload the -Series2- one. The -Gen04- one will not do anything for you.

If you dbloaded them according to those guidelines, it should work.

[tas3086]

Has anyone else had issues with 6.3f After I upgraded an HR10 each time I hit Messages and Setup the machine reboots. Machine seems to be stable other than that but that is an issue.

Mine works fine. The only problem that I had was with ip_tables.

[Barefooter]

Thanks for the info. I think I will give it another try.

Does anyone know how to do a dbunload. I did a search and came up blank.

[samhammer]

Thanks CrashHD

From another thread I was sending the Gen04 files rather than the Series2 files.

Now they show up in SWSystem.

Now to remember how to do the update.

Sam



Most likely, you missed one, or got them out of order.

you need to dbload:

GZcore
GZhpk-Series2
GZkernel-Series2
utils

Last you need to dbload the correct swsystem-xxxxxxxxxxxx file

The other 4 files must be dbloaded on the system (in any order)before swsystem. It must be last. Also, any slices in which there are 2 of them, one with -Series2- in the slice name, and another with -Gen04- in the slice name, you will need to dbload the -Series2- one. The -Gen04- one will not do anything for you.

If you dbloaded them according to those guidelines, it should work.

[neo96789]

Here's a good one...

After numerous reboots during the sara conner chronicals on Fox (the thumbs up bug I with 6.3d I suspect), I decided to dive in and apply the 6.3f slices and upgrade...

Of course, since it's been forever since I did an update, I skipped the step where I verified my kernal version after dd'ing it into the appropriate partition..causing the 6.3f upgrade to end in boot-loops due to the unhacked kernel...

I pulled the drive and reset my bootpage values, and re-installed my hacked kernel (just to be safe) on the 6.3d partitions, and rebooted...I'm now back up and running...

However: (and here's the interesting part), my MFS info says I'm runing 6.3f?!

It seems clear to me that I'm actually running 6.3d, since the SwSystem Environment variable = 6.3d-01-2, and my hacks are all still in place, but when I look at tivoweb's info screen, it tells me Software system 6.3f-01-2-357, and when I try the upgrade again it reports that I'm already running 6.3f?

I did a search here and could not locate anyone else having this problem...any ideas on what I can do to fix this short of waiting for 6.3g?

Thanks!

[crashHD]

that looks as though your bootpage flip failed for one reason or another. It looks like your running 6.3f MFS on the 6.3d binaries(root partition).

Mount the filesystem on your alternate partition, and cat /etc/build-version. If it says 6.3f, then you probably just need to flip the bootpage back to that partition.

[tivo4mevo]

It sounds as if the upgrade succeeded, which means that MFS was updated to reflect 6.3f (check /State/ServiceConfig in MFS) and your sysinfo screen reports it as such.

From an MFS point of view, you are running 6.3f (even though you booted to you 6.3d root when you flipped your bootpage). Perhaps the easiest solution would be to set your bootpage to the 6.3f root and correct what is causing the reboot loop.

Reboot loops are typically caused by incompatibilities between your kernel and network drivers. To debug the problem you could start with a using a stock 6.3f kernel, run replace_initrd on it, and use all stock network drivers. Alternatively, you could just try removing your network adapter to see if that helps.

[crashHD]

A kickstart 52 should reinstall the 6.3f binaries on the alternate root partition. It gives you a brand new, unmodified, 6.3f root partition. That should address whatever is causing the problem. The only downside I can see is it would mean pulling the drive to rehack it when you are done.

[dburckh]

I upgraded and to this and now my network is gone. Really weird stuff. The Tivo sees the card (verified by the MAC). I got bash working from a serial connection and this is what I tried:

ifconfig eth0 netmask 255.255.255.0 192.168.2.202 up
route add default gw 192.168.2.1

The route add fails with no route to network. I believe this is because the network adapter doesn't come up. If I do an ifconfig, eth0 doesn't show up unless I specify it specifically e.g. ifconfig eth0.

Any ideas? It's likely I screwed something up, but I don't know what it was.

[neo96789]

OK, almost there...?

I got 6.3f to boot ...the problem was the ipchains command...after I replaced it with a shell script which did nothing, I could boot into 6.3f...

However, I'm also having a network issue. I can get the adapter bound to an IP address, but it won't talk on it. ...I also cannot seem to get serial bash working for some reason...(doesn't work for me under 6.3d either...so maybe it's cables/etc)...

I can view the console output, just can't seem to get a console connection.

Anyway, thanks very much for the pointers that got me this far!...I'm gonna keep plugging away at this network issue.. I'm using modules and drivers from my previous config (6.3d with 3.1.5 kernel)...so maybe there's some problem there...

In the mean time, am I taking a risk by running with 6.3d software and MFS 6.3f?....

Thanks again!

[crashHD]

Thanks CrashHD
From another thread I was sending the Gen04 files rather than the Series2 files.
Now they show up in SWSystem.
Now to remember how to do the update.
Sam

/tvbin/installSw.tcl

edit installSw.tcl. Find the reboot command, and change it to an "exit" command. Call installSw.tcl with "tivosh /tvbin/installSw.tcl 6.3f-continue_rest_of_version_number"

When it's done installing, it will exit, and you will still have your telnet access until you reboot. You can mount your new root partition and rehack it from here.

[samhammer]

Thanks CrashHD,

Got it all done. Running 6.3f with all hacks in place.
Thanks for the help

Sam

[Roland Deschain]

6.3f slicepack, includes all SD series2, HR10, and R10.

6.3f (ed2k://|file|6.3f(6.3f-01.2_DirecTV_Tivo_slices).tgz|29018999|02EBAB75807C4D5367A037885862D0C0|h=SNW6LXM4ZWHZAVGGZBJLDH3P3A JJT26O|/)

6.3f slicepack on rapidshare. read .nfo file included in tgz......

http://rapidshare.com/files/76862446/6.3f_6.3f-01.2_DirecTV_Tivo_slices_.tgz

For those throwbacks among us who prefer BitTorrent, I've put the 6.3f slices up on BTJunkie:

http://btjunkie.org/torrent/TiVo-6-3f-Slices/2908928f5eaf495403ed597d07d53f8b212bcc7b58ce

[shawnT]

dburckh

I think I am having the same problem.
I don't use MRV I just set the ipaddress in rc.sysinit.author. This meant when I upgraded to 6.2 I just renamed /sbin/dhclint and /etc/netfilter.enable. Now there is no dhclient so I just renamed netfilter.enable.
But it went into a reboot loop so I changed iptables.
cd /sbin
mv iptables iptables.old
echo -e '#!/bin/bash\nexit 0' > iptables
chmod 755 iptables

This solved the reboot problem, but something is not working. Lights are on but nobody's home. Not sure if it is the drivers or not.

[dburckh]

dburckh

I think I am having the same problem.
I don't use MRV I just set the ipaddress in rc.sysinit.author. This meant when I upgraded to 6.2 I just renamed /sbin/dhclint and /etc/netfilter.enable. Now there is no dhclient so I just renamed netfilter.enable.
But it went into a reboot loop so I changed iptables.


This solved the reboot problem, but something is not working. Lights are on but nobody's home. Not sure if it is the drivers or not.

Thanks for the shout out, but I knew about the iptables issues.

I found a work around, so I'm good for now. The problem appears to have been related to a conflict between my hacked up version of PVNet and 6.3f DHCP client and possibly my new router. My solution was to remove all networking from rc.sysinit.author, enable HME/HMO and setup a static IP in the UI. I could not get DHCP to work period (either PVNet or Tivo version).

[crashHD]

I could not get DHCP to work period (either PVNet or Tivo version).
What kernel are you using?

[neo96789]

OK, almost there...?

I got 6.3f to boot ...the problem was the ipchains command...after I replaced it with a shell script which did nothing, I could boot into 6.3f...

However, I'm also having a network issue. I can get the adapter bound to an IP address, but it won't talk on it. ...I also cannot seem to get serial bash working for some reason...(doesn't work for me under 6.3d either...so maybe it's cables/etc)...

I can view the console output, just can't seem to get a console connection.

Anyway, thanks very much for the pointers that got me this far!...I'm gonna keep plugging away at this network issue.. I'm using modules and drivers from my previous config (6.3d with 3.1.5 kernel)...so maybe there's some problem there...

In the mean time, am I taking a risk by running with 6.3d software and MFS 6.3f?....

Thanks again!


OK, Gotit...!

Firstly, the problem with the console port was Hyperterminal....switched to Putty and it worked right away.

Secondly, network problem was that I needed to use netconfig.tcl...and needed to reboot twice after setting the IP with it...looks like tivoapp was overriding or somehow hosing up my ipconfig command.

Anyway...thanks for the help!

[dburckh]

What kernel are you using?

Not a clue. :) I had a monte, but I backed out believing it caused my DHCP issue. I think it's a 7.2. To add insult to my injury, I went to a SATA DVD drive and I have no clue how to mount it under Linux or if I even can. Now that I'm up again, I have others kernels that I could copy over. I'm willing to try them if you have a suggested version.

[jt1134]

Not a clue.
cat /proc/version - link (http://www.dealdatabase.com/forum/showthread.php?t=36692)
...I have others kernels that I could copy over. I'm willing to try them if you have a suggested version.
A 3.1.5 kernel won't work on 6.3x with DHCP unless you load af_packet.o manually. A 7.2.2-K-whatever kernel has DHCP functionality built in.

[crashHD]

Second that. I get the same DHCP results with 6.2a and those kernels.
3.1.5--no go
7.2.whatever.--DHCP functional

[dburckh]

Linux version 2.4.20 (build@buildmaster50) (gcc version 3.3.4) #1 Tue Feb 14 20:55:02 PST 2006
I think that means 7.2.x

[crashHD]

that looks right.

[jt1134]

Linux version 2.4.20 (build@buildmaster50) (gcc version 3.3.4) #1 Tue Feb 14 20:55:02 PST 2006
I think that means 7.2.x

Yep. DHCP should work.
DVD_TiVo#[/] $ mfs_dumpobj -r /State/Network
NetConfigIpParams 5475/11 {
AutoConfigEnabled[21]=0
DnsServers[20]=192.168.1.1
GatewayAddress[17]=192.168.1.1
HostAddress[16]=192.168.1.35
SubnetMask[18]=255.255.255.0
}
NetConfig 5475/10 PRIMARY {
Version[1]=2
Default[20]=1
HardwareAddress[17]=0 80 182 0 48 96
IpParams[18]=5475/11
NetworkMedium[16]=4
IndexPath[4]=/State/Network/004:5475
}
The above is set to static ip. Change AutoConfigEnabled to 1 to set to DHCP. Or you could use the hmo/hme patches to get the network setup UI screens.

[dburckh]

Might be my new router. Maybe the Tivo doesn't like it.

I picked up a TRENDnet TEW-633GR. Draft 2 Wireless N, Jumbo Frames, the whole enchilada. It feels a little too new though. I've been trying to bridge two. Didn't really work that well. Really slow as a bridge (<2 MBps). It get stable desktop speeds of >6 MBps. My old Linksys 350N would do up to 9 MBps, but it was really unstable.

Anywho, I'm OK with the static config. for now. Thanks for all the help.

[SpoonsJTD]

I'm also having trouble with getting DHCP to work on 6.3f. Setup:
HR10-250
6.3f
I'm using the iptables -F in rc.sysinit.author.

I have another HR10-250, with a nearly identical setup with same rc.sysinit.author, hacks, etc. except it is running 6.3e and it does DHCP fine.

Does it have something to do with dhclient being removed? Is anyone successfully using DHCP with 6.3f?

[jt1134]

I just tried to change my HR10 to DHCP via the menus and got an error N24, stating the network couldn't be found. Don't really feel dragging out my serial cable so I'm not gonna push it. Maybe you could force the AutoConfigEnabled variable to 1 and see if it works.

-edit- just flipped my bootpage to point back at my 6.3e root, rebooted, and fired up DHCP no problem. So the issue is obviously with the changes made to the root fs.

[SpoonsJTD]

dhclient is the app that *nix systems use to get DHCP leases, I'm thinking it a bit too coincidental that it was removed and now DHCP isn't working. :) I'm going to try to copy over dhclient and dhclient-script from a working 6.3e to my 6.3f box and see if that allows DHCP to be enabled.

[jt1134]

Well, that was easier than I thought it would be. Just stick dhclient back in /sbin and dhclient-script back in /etc, and you'll be good to go.


Posting too close together. :eek: It works. :)

[SpoonsJTD]

Dang! You beat me to it. :) I was just getting ready to post the same findings. Copied the two over, DHCP is back. dburckh, do what jt1134 said to fix it. :)

[jt1134]

Dang! You beat me to it. :)
What do I win? :D

[SpoonsJTD]

What do I win? :D

A lifetime supply of IP addresses supplied by your very own dynamic host configuration protocol server!

[Cash value .0001 of a cent]

I gave up too quickly over the weekend when trying to get this working. I forgot and\or didn't realize that my other 6.3x box was at 6.3e, not 6.3f. My tivo-hacking is spotty, I have periods of intense passion (typically due to new versions, new hacks, etc.) followed by longer periods of 'it works and my wife is happy that I haven't broken it by tinkering'. This last round of 'tinkering' was a result of me buying a 'broken' HR10-250 on e-bay for cheap and 'fixing' it by re-imaging it (after diagnostics said the HD was fine).

[sstrell]

Should the 6.3f superpatch included earlier in this thread disable encryption (it seems like it should)? I have an SDDVR40 and included it as part of the Zipper. But ciphercheck still tells me that I'm encrypted. Did I miss something? When I tried to do the echo command manually because it didn't look like encryption was disabled, I got stuck during boot and had to use my serial connection to replace tivoapp.

Thanks,
Steve

[Rorschach]

Using a script like Zipper is not something we discuss here and the site warns not to mention zipper. Is there some disk image of 6.3f compatible with such a script available now? Go back and read the instructions. You've screwed up.

[sstrell]

Sorry. Didn't know that. Is the superpatch script from earlier for series 2 DirecTivos like mine or for other machines that have 6.3f?

Steve

[SpoonsJTD]

[post removed]

I misread the question. That patch should work for any DTivo running 6.3f.

[kfcrary]

question withdrawn

[flagmaster]

Ok, I have an HR-10-250 subbed and hacked and it was working just fine with the OTA and the MPEG2 hd sat stations. That is till about 1 week ago when it would seem that a change has been made to the stream coming over the dish now. Since then both of my hacked tivos have been rebooting like crazy.

I assume that since they now offer remote scheduling on this model, that my very-old but very stable system software 3.1.5f is no longer going to cut it.

I do NOT have a phone line, I do have ethernet and serial bash access.

My plan is to upgrade to 6.3f today. This is my plan of attack.

1. Back up my system for when I screw it up I have a fall back plan.
2. Download the 6.3f slice package and use winrar to extract the files I need for the HR-10-250.

-GZcore-115422384-2.slice
-GZhpk-Series2-115422619-2.slice
-GZkernel-Series2-115422594-2.slice
-swsystem-115422985-2.slice
-utils-115422371-2.slice

3. Ftp these up to the tivo in binary and put them in /var
4. Telnet in, go to /var and dbload each one in that order.
5. Go to /tvbin/installSw.tcl and edit out the reboot and change to exit.
6. type....

tivosh /tvbin/installSw.tcl 6.3f-01-2-357

-----

At this point if I understand the procedure correctly, it will install the new system software onto my drive and move the kernal and root to the new partitions by doing a page flip.

Since I put in the "exit" instead of the reboot command, I should be able to mount the new root and install all my hacks.

Dose this sound right to you guys?

[flagmaster]

Wait a sec here guys....

I did this on my tivo just for fun...

echo mls /SwSystem | tivosh

and got this result....

bash-2.02# echo mls /SwSystem | tivosh
Directory of /SwSystem starting at ''

Name Type FsId Date Time Size
---- ---- ---- ---- ---- ----
3.1.5f-01-2-357 tyDb 769788 03/25/05 16:16 700
6.4a-01-2-357 tyDb 3458636 05/24/08 06:56 780
ACTIVE tyDb 769788 03/25/05 16:16 700

bash-2.02#

-------------------------------------

6.4.a ?????????

Ummmm, should I just use that? Do we know the tivo app offsets for 6.4a?

[crashHD]

6.4.a ?????????

Ummmm, should I just use that? Do we know the tivo app offsets for 6.4a?

Came out almost 2 months ago (http://www.dealdatabase.com/forum/showthread.php?t=59161). All known offsets for 6.3f have been also found for 6.4a. Superpatch and bufferhack have been ported. 6.4a hacked is good to go.

[flagmaster]

Humm, ok

Can you give me the link to the echo commands for the 30sec skip and disable enc for 6.4a

Also, what actually is the bufferhack and why would I need it?

[jt1134]

Can you give me the link to the echo commands for the 30sec skip and disable enc for 6.4a

he already did.

[flagmaster]

Ok, sorry about that.

I have edited this reply to something more sane. I kinda lost it this morning.

[flagmaster]

ok, this is where things went bad... I did the installSitcl thing and it looked like it was going to work just perfect...then towards the end of the script I got this....


-------
create: /install/dev/ttyUSB1
create: /install/dev/ttyUSB2
create: /install/dev/ttyUSB3
create: /install/dev/ttyUSB4
create: /install/dev/gen-ioctl
symlink /install/dev/log
ERROR: Didn't find any device list files in /install/platform/utils/DeviceList/
while executing
"exec tivosh /var/utils//buildskeleton /install"
("eval" body line 1)
invoked from within
"eval exec tivosh $prefix/buildskeleton $installdir"
(file "/var/utils/updateroot" line 147)
child process exited abnormally
while executing
"exec /var/utils/updateroot /dev/hda /install /var/packages >&@ stdout"
(object "::inst" method "::Installer::installSoftware" body line 46)
invoked from within
"inst installSoftware $name $Inc::TC_INSTALL_DIR "
(file "/tvbin/installSw.itcl" line 182)
bash-2.02#

=========================

The only thing that comes to mind is that since I have not upgraded since 2003, maybe my installerswitcl script may itself be too old to handle this major upgrade?


The only other thing I can think to try is to restore my backup and hook up a phone line and let it download the data (assuming they still send data over the telephone lines for software upgrades)???

[jt1134]

exec /var/utils/updateroot /dev/hda /install /var/packages >&@ stdout

find this line in installSw.itcl and change it to : exec /var/utils/updateroot /dev/hda /install /var/packages $name >&@ stdout

you also might want to export SerialNumber=`/tvbin/crypto -gsn` before running the install script.

[flagmaster]

ok, thanks for that...got this this time...



in serial bash I getting this as it boots and reboots and reboot and rebootss...
=====
Start fan control

First temperature parameters set:

Terminal temp: 71

Critical temp: 62

Logging temp: 60

Target temp: 50

Lowest fan speed: 7

/tvbin/fancontrol is running in the background.

Starting TvLauncher

Waiting for launcher to start.

Launcher is running.

Scanning for phase2 repair scripts

Running boot Stage E_PreApplication scripts

Checking for database conversions

Found binding for the hdmi semaphore

PHOENIX GETS MORE CONTIGMEM: height=1349

Illegal read at 00000014

do_page_fault #2: sending signal 11 to convert-db(147)

$0 : 00000000 fefefeff 00000005 5eee8005 00050010 00050010 7fff67d8 7fff67d0

$8 : 02a90ba0 80808080 000000f0 00000001 00000001 8620fd30 00000001 00000000

$16: 00050010 00050010 7fff67d8 00000000 00000000 00000002 5eeec2d0 00010014

$24: 00000000 02a25fd8 02add8b0 7fff6708 00010012 0046a8b4

Hi : 00000000

Lo : 00000022

epc : 0046a8b4 Tainted: P

Status: a0019413

Cause : 00000008

800bfb90 800bfbac 800c3e74 800c41c8 800c605c

0046a8b4 0046a854 0046aee0 00d07f8c 009a9b78 009a98cc 009a95cc 009a9630

009a8e74 00711404 0069d918 0074c31c 006dfda0 0201455c

Tmk Fatal Error: Thread convert-db <147> strayed!

pc 0x46a8b4 status 0x80019413 cause 0x000008 bva 00000000 hi 00000000 lo 0x000022

R00 0x00000000 R01 0xfefefeff R02 0x00000005 R03 0x5eee8005

R04 0x00050010 R05 0x00050010 R06 0x7fff67d8 R07 0x7fff67d0

R08 0x02a90ba0 R09 0x80808080 R10 0x000000f0 R11 0x00000001

R12 0x00000001 R13 0x8620fd30 R14 0x00000001 R15 0x00000000

R16 0x00050010 R17 0x00050010 R18 0x7fff67d8 R19 0x00000000

R20 0x00000000 R21 0x00000002 R22 0x5eeec2d0 R23 0x00010014

R24 0x00000000 R25 0x02a25fd8 R26 0x7fff6140 R27 0x00000000

R28 0x02add8b0 R29 0x7fff6708 R30 0x00010012 R31 0x0046a8b4

Paste the following into a shell to get a backtrace...


bt -t /tvbin/tivoapp <<END_OF_BT

tcd 1

hpk Series2

build b-firefly-takehome @280575 2007.10.02-1221 release-mips []

pack 6.3f-01-2

read 0x00400000 /tvbin/tivoapp

read 0x02000000 /lib/libc.so.6

read 0x02200000 /lib/libm.so.6

read 0x02400000 /lib/libpthread.so.0

read 0x02600000 /lib/libutil.so.1

read 0x02800000 /lib/libtvutil.so

read 0x02a00000 /lib/libtmk.so

read 0x02c00000 /lib/libtvstructures.so

read 0x2aaa8000 /lib/ld.so.1

read 0x2ab04000 /lib/libhpkoss.so

read 0x2ab50000 /platform/lib/libhpkhl.so

read 0x2ac18000 /platform/lib/libhpkll.so

read 0x2ac5c000 /lib/libdl.so.2

read 0x2aca0000 /lib/libcdaudio.so

0x0046a8b4 0x0046a854 0x0046aee0 0x00d07f8c 0x009a9b78 0x009a98cc 0x009a95cc

0x009a9630 0x009a8e74 0x00711404 0x0069d918 0x0074c31c 0x006dfda0 0x0201455c

END_OF_BT


Tmk Fatal Error: Thread convert-db <147>: unexpected signal 11

flushing ide devices: hda

Restarting system.
=====

is this a bad hdd?

[jt1134]

pack 6.3f-01-2

I thought you were trying to install 6.4a.

[flagmaster]

ya, that ended badly. (almost needed stitches!!)

Could that edit you gave me for 6.4 be the problem that Im seeing now because I went with 6.3?

If so, can we fix it, if not, can you point me to where I can get 6.4 slices?

I can do this all over again, but I need the 6.4 slices.

I no longer have them----thus the almost trip the the hospital when I smashed that hard drive ticking me off!

[jt1134]

Could that edit you gave me for 6.4 be the problem that Im seeing now because I went with 6.3?
the installSw.itcl fix has nothing to do with 6.3/6.4. it's a generic fix required for updating from 3.1.5x.

If so, can we fix it, if not, can you point me to where I can get 6.4 slices?
they're not publicly available that I have seen.

it's very hard to know what's going on since you're giving such sparse hints of what steps you've taken. my guess is your root fs doesn't match up with mfs, so it crashes when a 6.3f tivoapp is presented with a ??? mfs.

[RandC]

it's very hard to know what's going on since you're giving such sparse hints of what steps you've taken. my guess is your root fs doesn't match up with mfs, so it crashes when a 6.3f tivoapp is presented with a ??? mfs.There is the MFS conversion that was to occur with the 6.3+ upgrade, that might have something to do with the error.
I would try changing the boot page to the other partition to see if it boots up.

[flagmaster]

I think the boot page did change from 4 to 7, since the logo screen is completely different, and the serial bash looks complete different as well.

Should the "instalSWitcl" handle the MSF conversion, or is that done some other way?

This is what I can do...

I have a .BAK file made with MFS backup using -F 9999. This backup file is about 1385MB long. When I do an MFS restore, it looks like it restore a complete working 315 system. I even have a now playing list. Of course non of the recordings are included in this back up, so....

MFS restore to the drive.
Boot the drive in the tivo, get the error code #51.
Run the 51killer. (works)
Hand delete each recording till there are "No programs"
Then FTP the 6.3a slices to the /VAR as binary.
DBLOAD each one.
Edit the instalSwitcl as stated above.
tivosh run it.

Thats where Im at now.

What else should I try?

[RandC]

Should the "instalSWitcl" handle the MSF conversion, or is that done some other way?IIRC the MFS conversion happens after the reboot.

What else should I try?
Did you modify the tivoapp before the reboot when you upgraded? if so then I would insert the original tivoapp and reboot.

If that did not work then install original vmlinux and reboot, let the system wipe /var and reinstall all files. If that works, then you are back to having to reinstall your hacks.

From the error message "convert-db(147)" and "bt -t /tvbin/tivoapp <<END_OF_BT" it looks like it goes to convert and there is an error ion the tivoapp file.

[flagmaster]

The tivoapp file from 6.3 has not been mod'ed but the one from 315 is mod'ed.
Should that matter?

[RandC]

build b-firefly-takehome @280575 2007.10.02-1221 release-mips []

pack 6.3f-01-2

read 0x00400000 /tvbin/tivoapp

read 0x02000000 /lib/libc.so.6

read 0x02200000 /lib/libm.so.6

read 0x02400000 /lib/libpthread.so.0

read 0x02600000 /lib/libutil.so.1

read 0x02800000 /lib/libtvutil.so

read 0x02a00000 /lib/libtmk.so

read 0x02c00000 /lib/libtvstructures.so

read 0x2aaa8000 /lib/ld.so.1

read 0x2ab04000 /lib/libhpkoss.so

read 0x2ab50000 /platform/lib/libhpkhl.so

read 0x2ac18000 /platform/lib/libhpkll.so

read 0x2ac5c000 /lib/libdl.so.2

read 0x2aca0000 /lib/libcdaudio.so

0x0046a8b4 0x0046a854 0x0046aee0 0x00d07f8c 0x009a9b78 0x009a98cc 0x009a95cc

0x009a9630 0x009a8e74 0x00711404 0x0069d918 0x0074c31c 0x006dfda0 0x0201455c

END_OF_BT


Tmk Fatal Error: Thread convert-db <147>: unexpected signal 11

flushing ide devices: hda

Restarting system.
=====I had to do a little digging for error messages I save, and I had one similar to this right up to the libcdaudio file. My notes show I replaced the vmlinux with the original and then everything booted.

Also I noticed " pack 6.3f-01-2", I thought you were upgrading to 6.4a since you had the files in MFS? :confused:

[RandC]

The tivoapp file from 6.3 has not been mod'ed but the one from 315 is mod'ed.
Should that matter?I'm a HR10-250 background only. I do not think there is a difference in the tivoapps since the patch locations are the same. I still would boot with a clean tivoapp, 6.3f according to the error, a different 6.3x or 6.4x would cause problems.

EDIT - yes there is a difference between the 3.1.5x (315) tivoapp and 6.3x tivoapps. I was thinking of the different versions of 6.3X-01-2-XXX.

[flagmaster]

I only remember a little bit about the file "vmlinux". I think I worked with it when doing some stuff with monte but Im not sure.

In any case, where can I get a fresh copy of it (incase mine is the wrong one or mod'ed) and where dose it go? Is this file that is installed as the kernal with the DD command?

[RandC]

I only remember a little bit about the file "vmlinux". I think I worked with it when doing some stuff with monte but Im not sure.

In any case, where can I get a fresh copy of it (incase mine is the wrong one or mod'ed) and where dose it go? Is this file that is installed as the kernal with the DD command?Yes, dd into /dev/hda3 and/or /dec/hda6.
Here is clean vmlinux.px from 6.3f slices.

[flagmaster]

Two questions...

should I dd this to both 3 and 6 now and see if it works...

And...

Once it dose work, then can I put the killinithd 315 kernal back on to do my hacks?

[flagmaster]

Ok, putting the orignal kernal on both 3 an 6 did not help at all. Same errors, same place.

This is what I have done..

Restored the 351 bak image, and now its in the process of doing a "Clear and delete Everything".

Once that is done I will check back here for updates and sugguestions.

[RandC]

Ok, putting the orignal kernal on both 3 an 6 did not help at all. Same errors, same place.

This is what I have done..

Restored the 351 bak image, and now its in the process of doing a "Clear and delete Everything".

Once that is done I will check back here for updates and sugguestions.get the 6.3f slices here (http://www.dealdatabase.com/forum/showpost.php?p=291874&postcount=17) if you do not have them, dbload them, change installSw.itcl reboot line and exec /var/utils/updateroot /dev/hda /install /var/packages $name >&@ stdout line. Run installSw.itcl and copy hack and reboot. With all the trouble you have had you might just let the upgrade go thru completely, then after all is well pull the drive (unfortunately) to install kernel and hacks.

[flagmaster]

When ever I try to do a clear and delete everything, it says it will take 1 hour, but it NEVER finishis---EVER.

[RandC]

When ever I try to do a clear and delete everything, it says it will take 1 hour, but it NEVER finishis---EVER.If you have a working backup then I would not do a clear and delete everything. Just dbload slices and go. IIRC on a clear and delete you have to connect to DVR service via phone line to get back the recording capability.

[flagmaster]

Ok, still trying to clear and delete everything, aborting on that now,
restoring a working backup of 315.

What is "IIRC"?

[RandC]

What is "IIRC"?If I Remember Correctly

[flagmaster]

The clear and delete everything didnt work, I let it go for 4 hours with no hit of it getting done. Red led blinks as it should, but the serial bash shows no progress, and I dont think it would ever finish.

I have pulled the drive, restored a working 315 and ran the killer51.
I have the unit hooked to a phone line, and its downloading now. This will prolly take a while.

I will report back soon.

[flagmaster]

Well, the phone call to the system didnt download any new system software. I guess Im gonna have to do this manually.

ftp to var:

GZcore-115422384-2.slice.bin
GZhpk-Series2-115422619-2.slice.bin
GZkernel-Series2-115422594-2.slice.bin
swsystem-115422985-2.slice.bin
utils-115422371-2.slice.bin

dbload each one of these files..

edit the /tvbin/instalSwitc

execut it.

[flagmaster]

@#$@#$$@#!!!
@#%#$@!!
2#$@$%!!!
@#$@#!!
@@#%@%@#%@!

Im done.

I tried it with a fresh kernel on hda6, and a 315hacked kernal on hda6.
Both tries give the same result...

I tried both kernels with every .BAK I have access to, and I just cant make it take.

But at least the instalitcl works without any errors now.

I even tried 3 different 250 gig hard drives.

Im lost and my wife is going to kill me.

[jt1134]

using a virgin 6.3f kernel with any software other than 6.3f is a BAD idea. The initrd will nuke every file on the root fs since they don't match the initrd's stored hashes. restore 3.1.5x, fix installSw.itcl with only the $name fix, install 6.3f and make a backup of it before doing anything else. then at least you'll have a newer backup.

[flagmaster]

When I do the name$ fix and install 6.3 it works just fine, but when I reboot, I get this in the serial bash...

----
First temperature parameters set:

Terminal temp: 71

Critical temp: 62

Logging temp: 60

Target temp: 50

Lowest fan speed: 7

/tvbin/fancontrol is running in the background.

Starting TvLauncher

Waiting for launcher to start.

Launcher is running.

Scanning for phase2 repair scripts

Running boot Stage E_PreApplication scripts

Checking for database conversions

Found binding for the hdmi semaphore

PHOENIX GETS MORE CONTIGMEM: height=1349

Illegal read at 00000014

do_page_fault #2: sending signal 11 to convert-db(145)

$0 : 00000000 fefefeff 00000005 5eee8005 00050010 00050010 7fff67d8 7fff67d0

$8 : 02a90ba0 80808080 000000f0 00000001 00000001 7fd7fa48 00000001 00000000

$16: 00050010 00050010 7fff67d8 00000000 00000000 00000002 5eeec2d0 00010014

$24: 00000000 02a25fd8 02add8b0 7fff6708 00010012 0046a8b4

Hi : 00000000

Lo : 00000022

epc : 0046a8b4 Tainted: P

Status: a0019413

Cause : 00000008

8001e9fc 8001eac0 80022bb4 80022d74 80024db8 0046a8b4

0046a8b4 0046a854 0046aee0 00d07f8c 009a9b78 009a98cc 009a95cc 009a9630

009a8e74 00711404 0069d918 0074c31c 006dfda0 0201455c

Tmk Fatal Error: Thread convert-db <145> strayed!

pc 0x46a8b4 status 0x80019413 cause 0x000008 bva 0x81dbfde8 hi 00000000 lo 0x000022

R00 0x00000000 R01 0xfefefeff R02 0x00000005 R03 0x5eee8005

R04 0x00050010 R05 0x00050010 R06 0x7fff67d8 R07 0x7fff67d0

R08 0x02a90ba0 R09 0x80808080 R10 0x000000f0 R11 0x00000001

R12 0x00000001 R13 0x7fd7fa48 R14 0x00000001 R15 0x00000000

R16 0x00050010 R17 0x00050010 R18 0x7fff67d8 R19 0x00000000

R20 0x00000000 R21 0x00000002 R22 0x5eeec2d0 R23 0x00010014

R24 0x00000000 R25 0x02a25fd8 R26 0x7fff6140 R27 0x00000000

R28 0x02add8b0 R29 0x7fff6708 R30 0x00010012 R31 0x0046a8b4

Paste the following into a shell to get a backtrace...


bt -t /tvbin/tivoapp <<END_OF_BT

tcd 1

hpk Series2

build b-firefly-takehome @280575 2007.10.02-1221 release-mips []

pack 6.3f-01-2

read 0x00400000 /tvbin/tivoapp

read 0x02000000 /lib/libc.so.6

read 0x02200000 /lib/libm.so.6

read 0x02400000 /lib/libpthread.so.0

read 0x02600000 /lib/libutil.so.1

read 0x02800000 /lib/libtvutil.so

read 0x02a00000 /lib/libtmk.so

read 0x02c00000 /lib/libtvstructures.so

read 0x2aaa8000 /lib/ld.so.1

read 0x2ab04000 /lib/libhpkoss.so

read 0x2ab50000 /platform/lib/libhpkhl.so

read 0x2ac18000 /platform/lib/libhpkll.so

read 0x2ac5c000 /lib/libdl.so.2

read 0x2aca0000 /lib/libcdaudio.so

0x0046a8b4 0x0046a854 0x0046aee0 0x00d07f8c 0x009a9b78 0x009a98cc 0x009a95cc

0x009a9630 0x009a8e74 0x00711404 0x0069d918 0x0074c31c 0x006dfda0 0x0201455c

END_OF_BT


Tmk Fatal Error: Thread convert-db <145>: unexpected signal 11

flushing ide devices: hda

Restarting system.

--------

This is with 315killed kernel on both 3 and 6.

I even did the full upgrade, and dd'ed the 315 to the new active partition before the reboot, no difference.

[pentium101]

Well, the phone call to the system didnt download any new system software. I guess Im gonna have to do this manually.

ftp to var:

GZcore-115422384-2.slice.bin
GZhpk-Series2-115422619-2.slice.bin
GZkernel-Series2-115422594-2.slice.bin
swsystem-115422985-2.slice.bin
utils-115422371-2.slice.bin

dbload each one of these files..

edit the /tvbin/instalSwitc

execut it.I'm not sure if this is related to your problem, but are you making sure that you are dbloading the swsystem slice as the final slice?

[flagmaster]

Tried it your dbload order, same result, thanks tho.

[flagmaster]

A good friend let me make a new MFS backup of tivo that was already running at 6.4a.
That seems to have done the trick for me, Im back up and running.

I think there is a minor glitch with the backup files that I have of 315, not only will it fail to take the upgrade to 6.3 or 6.4 it also fails to ever complete a "Clear and Delete Everything" zap.

Its not a big deal at this point. I would like to know why, it would make the sleep a little better, but the bottom line is Im back up and running.

Thanks for all your help everyone.

[RandC]

Im lost and my wife is going to kill me.
That seems to have done the trick for me, Im back up and running.Good to hear you got it fixed before your wife went to jail. :D

[flagmaster]

Oh Man...

Just when I thought I was free and clear I have run into another small problem.

The image my friend gave me was from a different zip code and city then the one I live in. His image was from the Indianapolis. Indiana. I live in the same State, but my market is South Bend, Indiana.

I went in a made the changes to the zip code, and area code. It took them just fine. Then figured life was good.

Well I ran into a small problem...

It would seem that when I go into the Setup, Channles, it shows all the indy local hd stations, as well as my local south bend ones. I unchecked the indy ones (too far away to them over the air) and checked my local SB ones.

This seemed to take just fine, but now when I pick programs to record, it only shows the SAT channels with those show titles.

For example....

if I type in "Hell" it dose show the tv program on the side "Hell's Kitchen" but it only shows it on channel 250 (REAL) and not on my local Fox station.

Same with all other shows, only SAT channels show up.

I did a clear and delete everything that that seemed to work, it took me to the guided setup. I put in my zip and area code. It re-found my local channels and I can even watch live tv on them.

But when I try to do a Pick programs to record, it tells me that I need to make a phone setup call.

I do this, and it can only get to.....

preparing to call --- ok
dialing --- ok
connecting --- Fail while negotiating.

I tried dozens of times, and it keeps failing here.

This is a clip of serial bash---I think the problem has to do with the way the network was setup for USB, because it looks like the ppp connection is going just fine. It looks as tho there is a routing error.

Can this do its local call via usb? If so, what do I need to set up in the phone options?

------

Using interface ppp0
Connect: ppp0 <--> /dev/cua1
sent [LCP ConfReq id=0x1 <asyncmap 0x0> <magic 0x254098f3> <pcomp> <accomp>]
rcvd [LCP ConfReq id=0x1 <mru 1524> <asyncmap 0xa0000> <pcomp> <accomp>]
sent [LCP ConfAck id=0x1 <mru 1524> <asyncmap 0xa0000> <pcomp> <accomp>]
rcvd [LCP ConfAck id=0x1 <asyncmap 0x0> <magic 0x254098f3> <pcomp> <accomp>]
sent [IPCP ConfReq id=0x1 <addr 0.0.0.0> <compress VJ 0f 01>]
rcvd [IPCP ConfReq id=0x1 <compress VJ 0f 01> <addr 63.3.17.130>]
sent [IPCP ConfAck id=0x1 <compress VJ 0f 01> <addr 63.3.17.130>]
rcvd [IPCP ConfNak id=0x1 <addr 63.22.110.1>]
sent [IPCP ConfReq id=0x2 <addr 63.22.110.1> <compress VJ 0f 01>]
rcvd [IPCP ConfAck id=0x2 <addr 63.22.110.1> <compress VJ 0f 01>]
local IP address 63.22.110.1
remote IP address 63.3.17.130
Terminating on signal 15.
sent [LCP TermReq id=0x2 "User request"]
rcvd [LCP TermAck id=0x2]
Connection terminated.
Connect time 3.1 minutes.
Sent 48 bytes, received 42 bytes.

-------

[flagmaster]

I got figured out...

My pal had a route add statement in the rc.sysinit.author that blocked all ip traffic to tivo's main address. I temp remarked it out, and now my call went right thru, and looks like things are good again.

(I need to hide the gun).

[PlainBill]

I got figured out...

My pal had a route add statement in the rc.sysinit.author that blocked all ip traffic to tivo's main address. I temp remarked it out, and now my call went right thru, and looks like things are good again.

(I need to hide the gun).

Ordinarily I would add "Or at least stop fooling around with stuff you don't understand". But you did a good job of troubleshooting this last problem.

PlainBill

[kfcrary]

I have a basic question that I'm embarassed to ask, since I'm sure the information is already here somewhere, but I've been unable to uncover it.

I have an old Series 2 DTivo (a Hughes SD-DVR, those are great). I'm experimenting with updating from 6.2a to 6.3f but I've run into an unexpected obstacle. It seems that killhdinitrd (0.9.3) doesn't work on 6.3f's kernel.

All the talk in this thread seems to assume that the initrd check is already beaten. How is that done? Is there an older kernel that can run 6.3f, is everyone just using PROM mods, or is there some other technique I've missed? (I can't bear to go back to monte now.)

Thanks for any help you can offer.

(In case you're curious: I'm experimenting with 6.3f because I've had persistent problems with MRV (I'm guessing they're due to my local network architecture), but MFS_FTP always works fine, so I wanted to try out MovieLoader. I'd be willing to go with 6.4a if that turned out to be easier; I'm using 6.3f because I already had the slices lying around. And yes, I'm experimenting on a copy.)

[PlainBill]

This is a common question. Killhdinitrd works by exploiting a fortuitous arrangement of bytes in the stock kernel. Not all kernels are exploitable.

The original kernel supplied with 3.1.5 is exploitable; also the kernel supplied with 7.2.2-oth.k1-01-2 can be exploited. The latter is preferred; it is closer to the kernel supplied with 6.3f

PlainBill

[kfcrary]

Aha! That worked great, thanks!

[tvtyme]

...

I have an old Series 2 DTivo (a Hughes SD-DVR, those are great). I'm experimenting with updating from 6.2a to 6.3f but I've run into an unexpected obstacle. It seems that killhdinitrd (0.9.3) doesn't work on 6.3f's kernel.

All the talk in this thread seems to assume that the initrd check is already beaten. How is that done? Is there an older kernel that can run 6.3f, is everyone just using PROM mods, or is there some other technique I've missed? (I can't bear to go back to monte now.)

Thanks for any help you can offer.

(In case you're curious: I'm experimenting with 6.3f because I've had persistent problems with MRV (I'm guessing they're due to my local network architecture), but MFS_FTP always works fine, so I wanted to try out MovieLoader. I'd be willing to go with 6.4a if that turned out to be easier; I'm using 6.3f because I already had the slices lying around. And yes, I'm experimenting on a copy.)
Since you already have the v6.3f slices, just slice upgrade, dd the killhdinitrd'd 3.1.5 kernel (from the $5 PTVupgrade CD) to your new kernel partition, and reapply your hacks on the new root partition. Be advised - you'll lose MRV because the code is missing in v6.3x and 6.4x. MovieLoader is the replacement for it.

L8r