OS/2, eCS & ArcaOS - Technical > Storage

How long is a HPFS CHKDSK?

(1/3) > >>

Meeko:
2 hours and no sign of progress. VirtualBox's disk activity icon shows nothing, so I doubt the CHKDSK will ever finish. The volume is 30 gigs with only 197 megs in use, and the host system has a i7 CPU. I was using Danis506 to break the 2 gig barrier of IBM1S506.ADD. There seems to be no such thing as a fsck.hpfs for Linux or I would have used it. Since I was writing to a different volume when OS/2 locked up, can I just force OS/2 to mount it anyway or clear the dirty bit using dd in Linux? BTW, it is not the boot volume.

ivan:
The last time I did a chkdsk on a 20GB HPFS partition it took about 90 seconds.

May I ask why you want such a large HPFS partition in the first place?  Our standard partitioning on all our OS/2 machines and VMs is to have a 1GB HPFS boot partition and then several JFS partitions for such things as programs, data and storage.  On our test machine we have several 500GB JFS partitions created using the IBM JFS that appeared in one of the later fixpacks and with WSeB.

I just started up the test machine - it has auto chkdsk on all partitions - and it took 22 seconds from power on to full desktop (it has three 1TB disks and one 500GB disk - boot partition is 2GB HPFS the other six partitions are JFS).

eirik:
Erich,

from your post it appears you are running OS/2 // eCS in a Virtual Box.  Disk checking for virtual machines is usually not required (the host system does that job as I have understood matters).  Hence, abort (if it is still running), and just proceed.  Anyone with more expertise on the Virtual Box (I run the old Virtual PC with eCS as the host system and XP on the VM, and there the XP disk checking is of no importance).

Eirik

ivan:
Eirik,

The host system only looks after the virtual disk container file NOT the disks contained within that file.

If the partition within the VM becomes corrupted then the OS of that VM must be used to do the chkdsk.  In the case in question it could well be a constrained memory issue with such a large partition which could lead to a format and re-install with a better partition layout.

Roderick Klein:

--- Quote from: Eirik Romstad on February 15, 2015, 01:59:55 pm ---Erich,

from your post it appears you are running OS/2 // eCS in a Virtual Box.  Disk checking for virtual machines is usually not required (the host system does that job as I have understood matters).  Hence, abort (if it is still running), and just proceed.  Anyone with more expertise on the Virtual Box (I run the old Virtual PC with eCS as the host system and XP on the VM, and there the XP disk checking is of no importance).

Eirik

--- End quote ---

From what I can tell Windows (when not shut down properly) tends to enter into a CHKDSK a lot less then when OS/2 is not shut down quickly.
With Linux and OS/2/eCS a chkdsk is required, it does not matter if it runs on native hardware or in a virtual machine.
Especialy with a file system such as HPFS (which is not a journalling file system), you can not be certain in what state a file system is in.
So you need to run a CHKDSK.

As for HPFS and the maxium size of the HPFS partition. In theory its 65 gigabyte but all to often have I seen HPFS volume bigger 15 to 20 GB run into a trouble running a CHKDSK. Simply because it was never that well tested with such large volumes.

So its either use JFS with OS/2, and if you only have HPPS (such as Warp), then  go with HPFS.
Do your self a favor and never use FAT to install Warp 4 on in my experience it can blow up the quickests of all file systems (FAT/HPFS/JFS).

As Erich posts that his chdkdsk hangs. You will need to boot from floppies and the CD and then run a chkdsk. It could be the HPFS file system has received just a blow the CHKDSK can not even start. But thats hard to tell from a distance if the scenario I just described is happening on your system.

Roderick Klein

Navigation

[0] Message Index

[#] Next page

Go to full version