Is there a kernel compatibility tool for Puppy?
Is there a kernel compatibility tool for Puppy?
Are there any diagnostic tools to verify/troubleshoot the choices made by Puppy boot loader as to kernel modules suitable for various CPU configs? Although a version loads and seems to work OK (Xenial7.5) when it reboots, inode errors are always found.
That is not the kernel.
There are programs that will try to fix this kind of problem.
What format is this partition?
Fat32, ntfs, ext 2 or 3 or 4?
Need to know so you use the correct program.
Puppy has program for ext 2, or 3 or 4.
Fat32 or ntfs have to use Windows programs. Those are Windows formats.
How is Xenialpup 7.5 installed?
Full or frugal?
If frugal, using save file or save folder?
That is something wrong with the file system (format) on the partition that Xenialpup 7.5 is installed on.Inodes contain information such as ownership, permissions etc. of files. Each file has an inode-number which you can view using the ls command (ls -i). So basically inode errors are errors in the filesystem.
There are programs that will try to fix this kind of problem.
What format is this partition?
Fat32, ntfs, ext 2 or 3 or 4?
Need to know so you use the correct program.
Puppy has program for ext 2, or 3 or 4.
Fat32 or ntfs have to use Windows programs. Those are Windows formats.
How is Xenialpup 7.5 installed?
Full or frugal?
If frugal, using save file or save folder?
The things they do not tell you, are usually the clue to solving the problem.
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
Pup-Sysinfo->Mainboard->CPUany diagnostic tools to verify/troubleshoot the choices made by Puppy boot loader as to kernel modules suitable for various CPU configs
Gives a lot of info about the CPU and what it is doing.
Most kernels are configured to load whatever is needed to fully support all features of most of the common CPU's.
You got a specific CPU feature you do not think has been activated?
The things they do not tell you, are usually the clue to solving the problem.
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
No immediate crashes, but minor bugs in applications that other people don't see, and memory issues? over time. Will eventually hang if idled long enough.
Intel website implies my "Centrino Duo" is not quite the same as "Celeron M" that Sysinfo reports.
Inode errors are reported when loading the .2fs save file (which is on an NTFS partition). If there is no save file, there are no errors reported.
The FSCK was fixing the errors. Except until recently. A necessary driver (which was on the CD) was "not found", then upon further reboot, there was no desktop at all, although the boot was still "successful" and something was running.
(Have changed to .3fs save file, which is supposed to correct more errors.)
Was thinking of some tool that exercises every system call.
Intel website implies my "Centrino Duo" is not quite the same as "Celeron M" that Sysinfo reports.
Inode errors are reported when loading the .2fs save file (which is on an NTFS partition). If there is no save file, there are no errors reported.
The FSCK was fixing the errors. Except until recently. A necessary driver (which was on the CD) was "not found", then upon further reboot, there was no desktop at all, although the boot was still "successful" and something was running.
(Have changed to .3fs save file, which is supposed to correct more errors.)
Was thinking of some tool that exercises every system call.
.2fs save file is going to have issues with getting corrupted.
It is actually a ext2 format (file system) inside the save file.
If pfix=fsck is placed in the boot loader menu entry.
The file system is checked for errors every time it boots.
Example entry for Grub4dos boot loader:
If the ntfs formatted partition is not freshly formatted or ever used by Windows.
It could be fragmented.
That can cause issues, because the save file is not in one continuous location, but scattered in pieces.
Windows defrag program can be used to defrag the partition.
Affects .2fs, .3fs .4fs save files.
It is actually a ext2 format (file system) inside the save file.
If pfix=fsck is placed in the boot loader menu entry.
The file system is checked for errors every time it boots.
Example entry for Grub4dos boot loader:
Code: Select all
title Puppy bionicpup64 8.0 (sdc6/bionicpup6480uefi)
find --set-root uuid () 6b517a5d-03c6-4635-96c5-659b29518c84
kernel /bionicpup6480uefi/vmlinuz pdrv=6b517a5d-03c6-4635-96c5-659b29518c84 psubdir=/bionicpup6480uefi pmedia=atahd pfix=fsck
initrd /bionicpup6480uefi/initrd.gzIt could be fragmented.
That can cause issues, because the save file is not in one continuous location, but scattered in pieces.
Windows defrag program can be used to defrag the partition.
Affects .2fs, .3fs .4fs save files.
The things they do not tell you, are usually the clue to solving the problem.
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
Xenial obviously has pfix=fsck compiled onto the CD. Does the stuff written on the screen during boot get saved anywhere, or is there an option to boot step-by-step, where one could see each step?
In Windows, chkdsk usually caused more problems than it fixed if you used the "fix automatically" option. If two files got crosslinked, "fixing" ruined both, whereas trying each separately might find that one was OK, and a manual disk edit could disable the other. Fsck probably has report-but-don't-change also but that isn't an option since the options are already set.
Why do you think that fragmented files would cause errors? There isn't any reason for that. On a very empty NTFS partition with only the save.2fs file and its replacement being written, you wouldn't think there would be much fragmentation. (I thought one of the improvement of NTFS over FAT was that it looked for contiguous space for a given size file to be written?)
In Windows, chkdsk usually caused more problems than it fixed if you used the "fix automatically" option. If two files got crosslinked, "fixing" ruined both, whereas trying each separately might find that one was OK, and a manual disk edit could disable the other. Fsck probably has report-but-don't-change also but that isn't an option since the options are already set.
Why do you think that fragmented files would cause errors? There isn't any reason for that. On a very empty NTFS partition with only the save.2fs file and its replacement being written, you wouldn't think there would be much fragmentation. (I thought one of the improvement of NTFS over FAT was that it looked for contiguous space for a given size file to be written?)
If what was suppose to happen, always did, computers would not need fixing, ever! 
The save is more than just a simple file.
It is a file system inside a file.
You say you keep getting when it reboots, inode errors are always found.
I am telling you things that can cause the save to get corrupted.
I said to use Windows defrag program on a NTFS partition.
Have to run it in Windows.
To totally understand.
You boot from a CD install of Xenialpup.
The save is on the computers hard drive.
The hard drive partition, the save is on, is formatted NTFS.
Is all this correct?
The save is more than just a simple file.
It is a file system inside a file.
You say you keep getting when it reboots, inode errors are always found.
.If there is no save file, there are no errors reported
I am telling you things that can cause the save to get corrupted.
I said to use Windows defrag program on a NTFS partition.
Have to run it in Windows.
To totally understand.
You boot from a CD install of Xenialpup.
The save is on the computers hard drive.
The hard drive partition, the save is on, is formatted NTFS.
Is all this correct?
The things they do not tell you, are usually the clue to solving the problem.
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
When I was a kid I wanted to be older.... This is not what I expected
YaPI(any iso installer)
Yes.ou boot from a CD install of Xenialpup.
The save is on the computers hard drive.
The hard drive partition, the save is on, is formatted NTFS.
Is all this correct?
In /tmp, I found bootkernel.log, bootsysinit.log, and udevtrace.log which have useful stuff, I am sure, but only if you know what to look for. For instance, the bootsysinit says "No touchscreens detected", but this must be because it will be recognized as a mouse later?
The "/tmp/sfs_load" file doesn't mention any errors (now), but if there was a CD or hard disk (when saved to HD) problem, would errors show up in this log? What about errors in pupsave...2fs on HD?
There is a "/tmp/ntfsmnterr7294.txt" 0 byte file. Not clear whether there was an NTFS error or just the potential for an error, or what the number means. There are actually two NTFS partitions that are mounted during boot, but my data one is unmounted a second after the Xscreen comes up. Google implies "ntfs mount error" messaes should say more. Right now, there is no Windows available to exercise or fix the NTFS.
There is also a "/tmp/xerrs.log" with about 23,000 lines of "(ROX-Filer:6134): WARNING **: Attempt to paste empty selection!". Is this file an accumulation of all the errors since I created the new pupsave in the last week, or just this last boot of a few hours? Seems excessive. Web search was no help in explaining.
In /etc/rc.d/ I found rc.sysinit. This is where the screen messages like "echo -n "Making the filesystem usable..." >/dev/console" come from. So the answer to my question is those particular messages are not saved to a file? Would there be a way to create a log file for everything written to the screen by changing (and then unchanging after bootup is finished) the properties of /dev/console?
rc.sysinit also does the creation of fsckme.flg on the /mnt/home NTFS volume (not the linux filesystem), and a comment that it is deleted after successful shutdown. The actual call to fsck is somewhere in /initrd/init but the script is very complicated and I cannot find where the fsckme flag is used. In any case, the selected fsck program must be doing its own screen writes- to /dev/console?
Could there be a PUPMODE that pauses between operations?
Well, just had my first system hang (from Palemoon) and a power-off shutdown. Upon reboot, screen said e2fsck... recovering journal ...2 things about inodes, etc but it was all gone in a flash.
Now there is only a 0 byte file "ntfsmnterr7460.txt" instead of the other one, and xerrs.log only has 24 lines with no errors noted.
Sfs_load, again, doesn't mention anything about problems with the pupsave load.
I did get a Windows system to check the NTFS partitions: The Linux /home partition had no errors, another data partition (not mounted unless something is going to read/written) had undisclosed errors which chkdsk claims to have fixed.
Upon rebooting Puppy, there was no ntfsmnterr#### file in /tmp.
Then after mounting the data partition, "ntfsmnterr7461.txt" appeared.
So this error is coming from pmount and has no connection to the Linux kernel or the pupsave file.
So, back to the question, is there a way to verify or exercise all system calls instead of waiting for an application to hang up the system?
Upon rebooting Puppy, there was no ntfsmnterr#### file in /tmp.
Then after mounting the data partition, "ntfsmnterr7461.txt" appeared.
So this error is coming from pmount and has no connection to the Linux kernel or the pupsave file.
So, back to the question, is there a way to verify or exercise all system calls instead of waiting for an application to hang up the system?