9front - general discussion about 9front
 help / color / mirror / Atom feed
* [9front] hjfs type mismatch
@ 2024-10-31 23:30 william
  2024-11-01  2:03 ` qwx
  0 siblings, 1 reply; 6+ messages in thread
From: william @ 2024-10-31 23:30 UTC (permalink / raw)
  To: 9front

I switched over from cwfs to hjfs. Twice I seem to crash on cwfs and couldn't figure out why. 

So here I am with hjfs and i'm kind of lost. I was messing with werc under a user directory with 
the user having permissions and now I can't even see that folder and I have other issues on reboot

Here is a paste

/dev/sdC0: QEMU HARDDISK
/dev/sdC0/9fat   dos
/dev/sdC0/data
/dev/sdC0/fshjfs
/dev/sdC0/nvram
/dev/sdC0/plan9
/dev/sdD0: QEMU DVD-ROM
bootargs is (tcp, tls, il, local!device)[local!/dev/sdC0/fs -m 192 -a tcp!*!564]
hjfs: fs is /dev/sdC0/fs
keyfs: 0: no termination: [00][30]M[5d][5b][33]#[5d][4d][5b]M[64][5d][5b]#[62][5d][5b]M[33][5d]]
0 keys read in DES format
hjfs: type mismatch, dev /dev/sdC0/fs, block 139403, got raw, want indir, caller 0x59ae

init: starting /bin/rc
TronFS# ls
bin
lib
tmp
wercmacbook.tgz
TronFS# cd /usr/william
TronFS# ls
acme.dump
bin
lib
tmp
werc
wercmacbook.tgz
TronFS# cd werc
TronFS# ls
hjfs: type mismatch, dev /dev/sdC0/fs, block 138952, got raw, want dentry, caller 0x860d
ls: .: phase error -- type mismatch


Its like the user is dead/broken corrupt or something when I try to rcpu

rcpu -h tmryk -u william
lost connection

here is the log
tlssrv:  tls reports can't exec /bin/rc: file does not exist: '/bin/rc'

Thoughts?

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9front] hjfs type mismatch
  2024-10-31 23:30 [9front] hjfs type mismatch william
@ 2024-11-01  2:03 ` qwx
  2024-11-01  6:47   ` william
  0 siblings, 1 reply; 6+ messages in thread
From: qwx @ 2024-11-01  2:03 UTC (permalink / raw)
  To: 9front

On Fri Nov  1 00:30:14 +0100 2024, william@thinktankworkspaces.com wrote:
> I switched over from cwfs to hjfs. Twice I seem to crash on cwfs and couldn't figure out why. 

Crashing cwfs is not normal and something you're welcome to report,
along with some debug information.


> hjfs: type mismatch, dev /dev/sdC0/fs, block 138952, got raw, want dentry, caller 0x860d
> ls: .: phase error -- type mismatch
> 
> 
> Its like the user is dead/broken corrupt or something when I try to rcpu
> 
> rcpu -h tmryk -u william
> lost connection
> 
> here is the log
> tlssrv:  tls reports can't exec /bin/rc: file does not exist: '/bin/rc'
> 
> Thoughts?

Your file system is corrupt.

Which version of 9front and qemu are you using?  What format is the
virtual disk and how is it stored?  How are you powering the VM down?
Catastrophic failures with both filesystems seems suspicious.

Cheers,
qwx

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9front] hjfs type mismatch
  2024-11-01  2:03 ` qwx
@ 2024-11-01  6:47   ` william
  2024-11-01 13:03     ` qwx
  0 siblings, 1 reply; 6+ messages in thread
From: william @ 2024-11-01  6:47 UTC (permalink / raw)
  To: 9front

9front-10522.386.iso
its qcow2 or I installed it on an AWS ec2. i run qemu with tmux. I never power it off and it ran fine for about two weeks. No heavy usage. Just serves a single page. I copied the image back over to my system
to do further debugging and I just don't understand how or why. 

I have a linode server running fine for many years now but I can't seem to get something to run on AWS
as of late. 

Are there maintenance things I should be running like some sort of dump commands for daily weekly thing or?


Quoth qwx@sciops.net:
> On Fri Nov  1 00:30:14 +0100 2024, william@thinktankworkspaces.com wrote:
> > I switched over from cwfs to hjfs. Twice I seem to crash on cwfs and couldn't figure out why. 
> 
> Crashing cwfs is not normal and something you're welcome to report,
> along with some debug information.
> 
> 
> > hjfs: type mismatch, dev /dev/sdC0/fs, block 138952, got raw, want dentry, caller 0x860d
> > ls: .: phase error -- type mismatch
> > 
> > 
> > Its like the user is dead/broken corrupt or something when I try to rcpu
> > 
> > rcpu -h tmryk -u william
> > lost connection
> > 
> > here is the log
> > tlssrv:  tls reports can't exec /bin/rc: file does not exist: '/bin/rc'
> > 
> > Thoughts?
> 
> Your file system is corrupt.
> 
> Which version of 9front and qemu are you using?  What format is the
> virtual disk and how is it stored?  How are you powering the VM down?
> Catastrophic failures with both filesystems seems suspicious.
> 
> Cheers,
> qwx
> 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9front] hjfs type mismatch
  2024-11-01  6:47   ` william
@ 2024-11-01 13:03     ` qwx
  2024-11-01 17:40       ` william
  0 siblings, 1 reply; 6+ messages in thread
From: qwx @ 2024-11-01 13:03 UTC (permalink / raw)
  To: 9front

On Fri Nov  1 07:47:27 +0100 2024, william@thinktankworkspaces.com wrote:
> 9front-10522.386.iso
> its qcow2 or I installed it on an AWS ec2. i run qemu with tmux. I never power it off and it ran fine for about two weeks. No heavy usage. Just serves a single page. I copied the image back over to my system
> to do further debugging and I just don't understand how or why. 

You're using the 386 image, not the amd64 one?  I don't know much
about AWS or running VMs under it; there were bugs leading to file
corruption on 9front a while ago, but nothing I've heard of recently.


> Are there maintenance things I should be running like some sort of dump commands for daily weekly thing or?

No. If there is nothing else to go on, I'd blame the qemu on aws setup
causing some issue.  Check which image you're running and how.

Hope this helps,
qwx

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9front] hjfs type mismatch
  2024-11-01 13:03     ` qwx
@ 2024-11-01 17:40       ` william
  2024-11-01 21:45         ` qwx
  0 siblings, 1 reply; 6+ messages in thread
From: william @ 2024-11-01 17:40 UTC (permalink / raw)
  To: 9front

Does it matter which image I use? I can use both regardless of architecture and I often do although
I do notice a performance issue if it doesn't match the host OS.

Quoth qwx@sciops.net:
> On Fri Nov  1 07:47:27 +0100 2024, william@thinktankworkspaces.com wrote:
> > 9front-10522.386.iso
> > its qcow2 or I installed it on an AWS ec2. i run qemu with tmux. I never power it off and it ran fine for about two weeks. No heavy usage. Just serves a single page. I copied the image back over to my system
> > to do further debugging and I just don't understand how or why. 
> 
> You're using the 386 image, not the amd64 one?  I don't know much
> about AWS or running VMs under it; there were bugs leading to file
> corruption on 9front a while ago, but nothing I've heard of recently.
> 
> 
> > Are there maintenance things I should be running like some sort of dump commands for daily weekly thing or?
> 
> No. If there is nothing else to go on, I'd blame the qemu on aws setup
> causing some issue.  Check which image you're running and how.
> 
> Hope this helps,
> qwx
> 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9front] hjfs type mismatch
  2024-11-01 17:40       ` william
@ 2024-11-01 21:45         ` qwx
  0 siblings, 0 replies; 6+ messages in thread
From: qwx @ 2024-11-01 21:45 UTC (permalink / raw)
  To: 9front

On Fri Nov  1 18:42:31 +0100 2024, william@thinktankworkspaces.com wrote:
> Does it matter which image I use? I can use both regardless of architecture and I often do although
> I do notice a performance issue if it doesn't match the host OS.

Of course it matters.  I'm assuming you're only running 386 and amd64
vm's, and to oversimply to the extreme, one is 32-bit, the other is
64-bit.  There's no multilib or whatever on 9front.  Another more
important difference is that amd64 is de facto the standard (followed
by arm64), and 386 is used very rarely -- meaning that there probably
are more bugs in 386 just because fewer people test and use it.
Furthermore I am very doubtful that aws or other providers are running
32-bit systems limited to less than 4gb of RAM etc., hence the
"performance hit" you're probably referring to when you're running a
386 vm on there.  In any case, now we don't know if it's problems
specific to amd64 or 386, or anything.

Cheers,
qwx

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2024-11-01 21:46 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-10-31 23:30 [9front] hjfs type mismatch william
2024-11-01  2:03 ` qwx
2024-11-01  6:47   ` william
2024-11-01 13:03     ` qwx
2024-11-01 17:40       ` william
2024-11-01 21:45         ` qwx

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).