GSoC/kolibrios-nvme-driver
2
0
Fork 0
mirror of https://git.missingno.dev/kolibrios-nvme-driver/ synced 2026-03-09 12:53:24 +00:00
Code Issues 2 Actions Activity

Stable system hang when trying to write to NVMe disk using script #6

New Issue
Closed
opened 2024-08-09 17:11:06 +00:00 by Burer · 9 comments
Burer commented 2024-08-09 17:11:06 +00:00
Copy Link

A very strange bug, that occurs only when trying to write data on drive using some sort of script.

For example, I can create, edit, copy-paste and delete files using Eolite freely, but when I try to compile C file and save executable in the same folder on NVMe drive - it just hangs up the whole system instantly. While when saving result on other drives, for example, TMP, all works just fine.

Will try to test it more, using different script and other write methods.

A very strange bug, that occurs only when trying to write data on drive using some sort of script. For example, I can create, edit, copy-paste and delete files using Eolite freely, but when I try to compile C file and save executable in the same folder on NVMe drive - it just hangs up the whole system instantly. While when saving result on other drives, for example, TMP, all works just fine. Will try to test it more, using different script and other write methods.
nvme_37ea6e2d16_virtualbox_7.png
154 KiB
nvme_37ea6e2d16_virtualbox_7.png
Burer changed title from Stable system hang when trying to write to NVMe disk to Stable system hang when trying to write to NVMe disk using script 2024-08-09 17:11:28 +00:00
Burer commented 2024-08-10 13:11:25 +00:00
Author
Copy Link

Update for commit 0f24000b9a, this Issue seems to be fixed.
But I guess we could use some other testing scenarios here before drawing definite conclusions.

Update for commit 0f24000b9a, this Issue seems to be fixed. But I guess we could use some other testing scenarios here before drawing definite conclusions.
nvme_0f24000b9a_virtualbox_2.png
120 KiB
nvme_0f24000b9a_virtualbox_2.png
ramenu commented 2024-08-10 15:44:13 +00:00
Member
Copy Link

I think this is a duplicate of #3 but I'm not sure.

I think this is a duplicate of https://git.kolibrios.org/GSoC/kolibrios-nvme-driver/issues/3 but I'm not sure.
Burer commented 2024-08-10 16:16:47 +00:00
Author
Copy Link

I think this is a duplicate of #3 but I'm not sure.

I am not sure either.
On previous commits, writing to NVMe drive with script caused a stable freeze, while on commit 0f24000b9a it worked fine a few times in a row.
While with deleted folder deletion - on commit 0f24000b9a, it froze first time and deleted folder without problems on second.
So I will try to test it a little bit more and provide additional information.

> I think this is a duplicate of https://git.kolibrios.org/GSoC/kolibrios-nvme-driver/issues/3 but I'm not sure. I am not sure either. On previous commits, writing to NVMe drive with script caused a stable freeze, while on commit 0f24000b9a it worked fine a few times in a row. While with deleted folder deletion - on commit 0f24000b9a, it froze first time and deleted folder without problems on second. So I will try to test it a little bit more and provide additional information.
ramenu commented 2024-08-10 16:35:31 +00:00
Member
Copy Link

If you could screenshot the screen when it freezes, that would be helpful. Thanks.

If you could screenshot the screen when it freezes, that would be helpful. Thanks.
ramenu commented 2024-08-10 17:04:23 +00:00
Member
Copy Link

@Burer, lol so this is hilarious..

Apparently the driver works without crashing when you write debug output to the screen, otherwise it doesnt. But anyways I think I've fixed the command wrapping issue now, so that's good. Will add into the next commit

@Burer, lol so this is hilarious.. Apparently the driver works without crashing when you write debug output to the screen, otherwise it doesnt. But anyways I think I've fixed the command wrapping issue now, so that's good. Will add into the next commit
Burer commented 2024-08-11 06:46:51 +00:00
Author
Copy Link

Update for commit 9ace6a4b52. I wasn't able to reproduce bag with deletion of this folder with debug output on screen, but got it first time when trying to delete this folder without debug output on screen.
But I also got the same freeze when copying it with debug output on screen, despite some part of it was really copied.
Interesting thing - when I created a bunch of copies of tcc folder in Linux and then tried to delete them (to create this .Trash-999 folder) I also got a bunch of errors like "file not found".

Update for commit 9ace6a4b52. I wasn't able to reproduce bag with deletion of this folder with debug output on screen, but got it first time when trying to delete this folder without debug output on screen. But I also got the same freeze when copying it with debug output on screen, despite some part of it was really copied. Interesting thing - when I created a bunch of copies of tcc folder in Linux and then tried to delete them (to create this .Trash-999 folder) I also got a bunch of errors like "file not found".
nvme_9ace6a4b52_virtualbox_3.png
107 KiB
nvme_9ace6a4b52_virtualbox_2.png
107 KiB
nvme_9ace6a4b52_virtualbox_1.png
154 KiB
nvme_9ace6a4b52_virtualbox_3.png nvme_9ace6a4b52_virtualbox_2.png nvme_9ace6a4b52_virtualbox_1.png
ramenu commented 2024-08-11 16:02:00 +00:00
Member
Copy Link

@Burer: Make sure you're not using corrupted data. It's possible on some of your runs that the NVMe driver didn't finish writing some stuff and now the data is corrupted or invalid. It happened to me a couple times.

But yes, I think this is the final serious issue. So I'll get started working on this.

@Burer: Make sure you're not using corrupted data. It's possible on some of your runs that the NVMe driver didn't finish writing some stuff and now the data is corrupted or invalid. It happened to me a couple times. But yes, I think this is the final serious issue. So I'll get started working on this.
ramenu commented 2024-08-11 18:39:27 +00:00
Member
Copy Link

@Burer, Okay I've tested this and I think it should be fixed in b0ead51d7f. Please try that. Thanks once again.

@Burer, Okay I've tested this and I think it should be fixed in https://git.kolibrios.org/GSoC/kolibrios-nvme-driver/commit/b0ead51d7fc95f38b36e91c90bb9b6cb2af22568. Please try that. Thanks once again.
Burer commented 2024-08-12 18:29:06 +00:00
Author
Copy Link

@Burer, Okay I've tested this and I think it should be fixed in b0ead51d7f. Please try that. Thanks once again.

Tested and posted results in discussion of #3.
As for this problem - it seems to be solved for me, I can't reproduce it in latest commits (starting from 0f24000b9a, I suppose).

> @Burer, Okay I've tested this and I think it should be fixed in https://git.kolibrios.org/GSoC/kolibrios-nvme-driver/commit/b0ead51d7fc95f38b36e91c90bb9b6cb2af22568. Please try that. Thanks once again. Tested and posted results in discussion of #3. As for this problem - it seems to be solved for me, I can't reproduce it in latest commits (starting from 0f24000b9a, I suppose).
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
No items
No Label
Milestone
No items
No Milestone
Assignees
Clear assignees
dunkaist ramenu Sweetbread
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: GSoC/kolibrios-nvme-driver#6
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?