Skip to contentSkip to conversation

TestPerlDBDmysql continues to fail

Buildroot Thomas Petazzoni 3 participants 5 messages Apr 22, 2019 – Apr 27, 2019

lore
Share & tools
lore
#0 lore raw

Hello Fran?ois,

The test TestPerlDBDmysql continues to fail in Gitlab CI:

https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668

Could you have a look and propose a fix for this issue ? I think it has been failing like this for quite some time, if I remember correctly.

Thanks a lot!

Thomas

Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
#1 lore raw

Le lun. 22 avr. 2019 ? 22:12, Thomas Petazzoni <[email protected]> a ?crit :

Quoted 10 lines

Hello Fran?ois,

The test TestPerlDBDmysql continues to fail in Gitlab CI:

https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668

Could you have a look and propose a fix for this issue ? I think it has been failing like this for quite some time, if I remember correctly.

I could reproduce the issue.
It is weird:
- the build step seems good, the image rootfs.cpio is created
- but the system does not boot, there is a kernel panic, see
https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668/artifacts/file/test-output/TestPerlDBDmysql-run.log

Outside the testing infrastructure, I could build an image with the module perl-dbd-mysql and run successfully:

$ perl -v
$ perl -MDBI -e 1
$ perl -MDBD::mysql -e 1

Fran?ois

Quoted 12 lines

Thanks a lot!

Thomas
--
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20190423/dc1adb46/attachment.html>

#2 lore raw
Quoted 5 lines · On Tue, 23 Apr 2019 09:44:33 +0200 Fran?ois Perrad <[email protected]>…

I could reproduce the issue.
It is weird:
- the build step seems good, the image rootfs.cpio is created
- but the system does not boot, there is a kernel panic, see
https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668/artifacts/file/test-output/TestPerlDBDmysql-run.log

I can reproduce the issue locally. If I take the same rootfs.cpio, and boot it with -M versatilepb and the kernel-versatile kernel image, it boots fine.

However that is strange because:

- The ARMv5 user-space that is being built should work fine under an ARMv7 machine

- All other Perl test cases are using the same base configuration (ARMv5) and booting under an ARMv7 Qemu machine + kernel, and they work fine.

Thomas

Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
#3 lore raw
Quoted 13 lines · Thomas Petazzoni
Quoted 5 lines · On Tue, 23 Apr 2019 09:44:33 +0200 Fran?ois Perrad <[email protected]>…

I could reproduce the issue.
It is weird:
- the build step seems good, the image rootfs.cpio is created
- but the system does not boot, there is a kernel panic, see
https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668/artifacts/file/test-output/TestPerlDBDmysql-run.log

I can reproduce the issue locally. If I take the same rootfs.cpio, and boot it with -M versatilepb and the kernel-versatile kernel image, it boots fine.

Quoted 8 lines

However that is strange because:

- The ARMv5 user-space that is being built should work fine under an ARMv7 machine

- All other Perl test cases are using the same base configuration (ARMv5) and booting under an ARMv7 Qemu machine + kernel, and they work fine.

The actual error is:

Unpacking initramfs...

Initramfs unpacking failed: write error

(it's quite a bit before the end of the file).

This rootfs is about 40MB, while the others are around 32MB. So I guess it just
doesn't fit in the 128MB given to the system.

Regards,
Arnout

#4 lore raw

Le sam. 27 avr. 2019 ? 01:26, Arnout Vandecappelle <[email protected]> a ?crit :

Quoted 30 lines · Thomas Petazzoni
Quoted 8 lines
Quoted 5 lines · On Tue, 23 Apr 2019 09:44:33 +0200 Fran?ois Perrad <[email protected]>…

I could reproduce the issue.
It is weird:
- the build step seems good, the image rootfs.cpio is created
- but the system does not boot, there is a kernel panic, see

https://gitlab.com/buildroot.org/buildroot/-/jobs/199339668/artifacts/file/test-output/TestPerlDBDmysql-run.log

Quoted 5 lines

I can reproduce the issue locally. If I take the same rootfs.cpio, and boot it with -M versatilepb and the kernel-versatile kernel image, it boots fine.

Quoted 8 lines

However that is strange because:

- The ARMv5 user-space that is being built should work fine under an ARMv7 machine

- All other Perl test cases are using the same base configuration (ARMv5) and booting under an ARMv7 Qemu machine + kernel, and they work fine.

The actual error is:

Unpacking initramfs...

Quoted 9 lines

Initramfs unpacking failed: write error

(it's quite a bit before the end of the file).

This rootfs is about 40MB, while the others are around 32MB. So I guess
it just
doesn't fit in the 128MB given to the system.

This test was initially written with Perl 5.26 which gives a rootfs a little smaller and allow to run successfully.

Fran?ois

Quoted 4 lines

Regards,
Arnout

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20190427/474f5fee/attachment.html>