Here is something for you all to bang your heads on, it is driving me insane.
A couple of weeks ago Odyssey gave me a GrimReaper and refused to launch. Weird I thought, and I don't even remember what I tried to do to make it work again, probably I made a fresh install, but it didn't work.
So I booted the spare SYS partition and voilá, it launched! My immediate thought was that my main System Partition has to have some issue, in fact when I booted from there Odyssey GrimReapers a launch.
I thought I'd do a fresh install of the System Partition and boy o boy it is really becoming a drag instead of a pleasant hobby thing to do. AmigaOS 4.1 Final Edition, then the updates, then being careful to not screw up with the radeon drivers, then install part of the Enhancer package. Modify the Kicklayout for the RadeonRX... Not Cool!
Long story short, Odyssey didn't want to launch when starting from the freshly installed partition.
So I musste have missed something.
At the end I replaced the SSD with a brand new one, created 4 partitions, SSD0 (Fresh install of OS4.1FEu2 with Enhancer stuff), SSD1 (a copy of the System partition that launches Odyssey), SSD2 (OS4.1FE u2 vanilla), SSD3 (Work).
SSD0 does not want to start odyssey. I cloned SSD1 to SSD0 and used CompareDirs to verify it is the same, icon files differ, appdir entries differ once I launch some apps, and some directories have different creation dates (even though I copied using the CLONE option). No success, Odyssey refuses to launch. I am double-clicking on the same "Odyssey" from the same drawer, there are no special setting in Odyssey, just the latest available version. It resides in the SSD3(Work) partition and starts when I boot the system from SSD1 or SSD2 but not SSD1.
I went ahead of me and tried the same with yet another drive, a new "conventional" sata HDD I had as spare. I removed all other drives to exclude any interference.
So any ideas where I can start to search for bugs? The kicklayout points to the kickstart modules in the system partition it is booting from, so no error there I guess.
In the CFE menu I choose the option to boot from the "amiga menu" and then the appropriate option for SSD0, 1, or 2.
To recap: The SSD0 partition is the same as the SSD1: partition (except some info file, but they were clone copied from SSD1, so no tool type should be different). Odysses launches when the system is started from SSD1 put grim Reapers when it is started from SSD0:
The original problem was Odyssey refusing to run after a Grim Reaper crash...but you could still boot into that SYS: partition? Correct?
I don't know why you're having trouble after a clean reinstall of everything, but the first problem could have been a corrupt Odyssey cache, cookie file, etc.
@walkero I did. nothing I can see. It really is exactly the same System if I am to believe the "copy clone" command of the OS (and I also tried the same using the lates "copy" from enhancer) and DirCompare.
Of course I took care to recreate the links in SOBJS: When I use Snoopy to start Odyssey, the System eventually crashes and there is even no output on serial. @tao
Quote:
The original problem was Odyssey refusing to run after a Grim Reaper crash...but you could still boot into that SYS: partition? Correct?
I don't know why you're having trouble after a clean reinstall of everything, but the first problem could have been a corrupt Odyssey cache, cookie file, etc.
When I start Odyssey from partition A it crashes, when I start it from partition B it doesn't crash (GrimReaper). Odyssey starts from exactly the same directory: "Work:Apps/Internet/Odyssey"
@nbache @vgappc
SmartFileSystem 1.293 (Enhancer)
@LiveForIt
It is an X1000 with CFE PAS-v2.0.30 AmigaOS 4.x OpenFirmware Bootloader V53.21
I am using version 50.3 from os4depot. To have total certainty, I downloaded it this morning, copied to the problematic partition, fully rebooted (restart computer)
TSSTcorp CDDVDW SH-223DB Removable CD-ROM or CD-RW or DVD not supported Netac SS YS2022100036554 Fixed hard disk ready Kingston 509483948948308D Fixed hard dis not installed (Linux)
@Tuvok I still try to check if you really boot from partition you select ( because with certain firmware + kickstart setup is possible, that you select one kickstart, which is loaded, but system after that booted from partition with highest priority, or from first partition when priorities are the same).
Can you please check some more variables? Your CFE vars are modified from default setting. MENU_X_COMMAND normally looks like:
Boot partition with the highest priority: MENU_0_COMMAND=setenv amigaboot_quiet Y ; boot -fs=amigafs ide0.0:amigaboot.of
Envoke menu for selecting AmigaOS boot partition: MENU_1_COMMAND=setenv os4_bootdevice "auto" ; boot -fs=amigafs ide0.0:amigaboot.of
it looks like your CFE have defined some additional variables, maybe something like like: ??? abootquiet=setenv amigaboot_quiet Y ??? os4bdauto=setenv os4_bootdevice "auto" ??? bootAFS=boot -fs=amigafs ??? abootof=ide0.0:amigaboot.of I am also not 100% sure what "$" means. In CFE Manual no "$" operator is mentioned, but probably it the same like in bash - envoke variable value.
Can you please check in CFE all environmental variables used in your menus? (abootquiet, bootAFS, $abootof, os4bdauto) Mainly is necessary check you have correctly substituted setenv os4_bootdevice "auto" in the case of MENU_1_COMMAND. With this correctly setted, AOS should boot from the same partition like kickstart was loaded.
There can be also second issue: If you have on your partition(s) file Kickstart/BootDevice it can override setenv os4_bootdevice "auto" setting. In case you use Kickstart/BootDevice files, please check if you have correctly setted the value of it, i.e.: On SSD0 partition should be SSD0 in Kickstart/BootDevice file, On SSD1 partition should be SSD1 in Kickstart/BootDevice file, on SSD2 partition should be SSD2 in Kickstart/BootDevice file.
Sorry, maybe all this is unnecessary, but symptoms of your booting error is very strange, and is possible that you boot AOS from different partition than kickstart.
AmigaOS3: Amiga 1200 AmigaOS4: Micro A1-C, AmigaOne XE, Pegasos II, Sam440ep, Sam440ep-flex, AmigaOne X1000 MorphOS: Efika 5200b, Pegasos I, Pegasos II, Powerbook, Mac Mini, iMac, Powermac Quad
Worth a look just in case. Can cancel that out then. In that case the Grim Reaper is the next best clue. Which can save a log. Full of lots of info but in most cases the stack trace usually is all that's needed to pin point an issue.
Hi, thank you for your replies and sorry for taking so much time to reply myself.
Quote:
it looks like your CFE have defined some additional variables, maybe something like like: ??? abootquiet=setenv amigaboot_quiet Y ??? os4bdauto=setenv os4_bootdevice "auto" ??? bootAFS=boot -fs=amigafs ??? abootof=ide0.0:amigaboot.of I am also not 100% sure what "$" means. In CFE Manual no "$" operator is mentioned, but probably it the same like in bash - envoke variable value.
I checked all the CFE variables, everything here works as ist should. As the two partitions are (or should be identical) I don't see what difference it would make if the system loads the kickstart from one partition or the other.
However, I double checked, when selection the CFE menu to select the boot partition, it really is the selected partition the system is booted from. Same goes for the "AmigaBoot-menu" it does boot from the selected partition.
And when the system has booted in the os, in the CLI/Shell I could confirm which system has booted.
Furthermore I disabled the partitions one at the time, to see if it makes any difference; other than I could really be sure the system had not booted from a different partition than the one chosen, there is none.
This is really so puzzling. I am tempted to just overwrite the partition and system and be good with it, but it really bothers me. If the two partitions are the same, why o why does one allow Odyssey to work and the other not?
It was asyncio.library. Somehow a very old version of the year 2002 sneaked into the system. Only thing that puzzles me, is that it works when starting from one partition, but doesn't from the other. I even went all the way to install the two system partition on two different empty SSDs with only on system partition on it and nothing else. I unplugged the other SSD, so it was 100% the system did start from the correct partition and nothing interfered.
Well I'll put this in the "mystery drawer" and let it rest, as I have no ideas left on how to find out the difference of the apparently identical partitions (tested with CompareDirs from os4depot).