[SOLVED] When archiso build.sh script, the mount for airootfs.img fail

Question

Solved, you can see the details down in this post https://bbs.archlinux.org/viewtopic.php … 6#p1481476. Basically I rebooted.
VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
As the subject states, when running ./build.sh there is a mount: work/mnt/airootfs: mount failed: Unknown error -1
looking through the script I think the command that is being ran is
mount /home/pingbuild/archlive/baseline/work/airootfs.img /home/pingbuild/archlive/baseline/work/mnt/airootfs
The mnt/airootfs directory does exist. I guess normally the directory is removed, but I am assuming the scripts stop with the error. Anyway if I run the command by hand I get the same error.
Is the airootfs.img not being created correctly?
Here is a snapshot of the std out. I could not get std err to go to curl/spurgen, so the actual error did not upload. I will paste by hand so you can see where it would have been.
[mkarchiso] INFO: Configuration settings
[mkarchiso] INFO: Command: prepare
[mkarchiso] INFO: Architecture: x86_64
[mkarchiso] INFO: Working directory: work
[mkarchiso] INFO: Installation directory: arch
[mkarchiso] INFO: Cleaning up what we can on airootfs...
[mkarchiso] INFO: Done!
[mkarchiso] INFO: Creating ext4 image of 32GiB...
Discarding device blocks: 4096/8388608 done
Creating filesystem with 8388608 4k blocks and 2097152 inodes
Filesystem UUID: df2e27ca-bdaf-4923-9568-f066c3f57b6c
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624
Allocating group tables: 0/256 done
Writing inode tables: 0/256 done
Writing superblocks and filesystem accounting information: 0/256 done
[mkarchiso] INFO: Done!
[mkarchiso] INFO: Mounting 'work/airootfs.img' on 'work/mnt/airootfs'
mount: work/mnt/airootfs: mount failed: Unknown error -1
[mkarchiso] INFO: Unmounting 'work/mnt/airootfs'
umount: work/mnt/airootfs: not mounted
Last edited by penright14 (2014-12-04 13:49:27)

Answer 1

djgera wrote:See https://bbs.archlinux.org/viewtopic.php?id=190477
Thanks, after looking at the post, I did not see that in the wiki.
My systemd is 217-8
I may have misunderstood what the post was accomplishing, but I run 'modprobe loop', then tried my mount same error. If I may ask a dumb question, once 'modprobe loop' has ran without any errors shown, will the loop be added until the next boot?
Here is my lsmod
Module Size Used by
mousedev 17272 0
ppdev 16782 0
coretemp 12820 0
hwmon 12930 1 coretemp
vmw_balloon 12658 0
crc32c_intel 21809 0
evdev 21544 1
mac_hid 12633 0
psmouse 107214 0
serio_raw 12849 0
pcspkr 12595 0
e1000 130280 0
battery 17452 0
vmwgfx 174313 1
ttm 73743 1 vmwgfx
drm_kms_helper 80934 1 vmwgfx
drm 259106 4 ttm,drm_kms_helper,vmwgfx
i2c_piix4 20857 0
i2c_core 50152 3 drm,i2c_piix4,drm_kms_helper
parport_pc 26351 0
parport 35749 2 ppdev,parport_pc
irda 113859 0
crc_ccitt 12347 1 irda
vmw_vmci 55463 0
intel_agp 17432 1
intel_gtt 17848 1 intel_agp
shpchp 35210 0
processor 27777 0
ac 12715 0
button 12953 0
sch_fq_codel 17343 2
ext4 501833 2
crc16 12343 1 ext4
mbcache 17171 1 ext4
jbd2 86417 1 ext4
sr_mod 21903 0
cdrom 47479 1 sr_mod
ata_generic 12490 0
pata_acpi 12771 0
sd_mod 44398 3
crc_t10dif 12431 1 sd_mod
crct10dif_common 12356 1 crc_t10dif
ata_piix 33592 0
mptspi 21940 2
scsi_transport_spi 27754 1 mptspi
mptscsih 26657 1 mptspi
mptbase 72945 2 mptspi,mptscsih
libata 181518 3 pata_acpi,ata_generic,ata_piix
scsi_mod 147543 6 scsi_transport_spi,libata,mptspi,sd_mod,sr_mod,mptscsih
atkbd 22254 0
libps2 12739 2 atkbd,psmouse
floppy 69118 0
i8042 18002 1 libps2
serio 18282 6 serio_raw,atkbd,i8042,psmouse
Also
djgera wrote:
Please check if loop kernel module is loaded. Also, check for any related kernel log messages.
truncate -s 1M coco.img
mkfs.vfat coco.img
mount coco.img /mnt
does this works?
Gives me the same results. Also I tried overwriting the vfat with etx4, same thing. This confused me. I know I have made rootfs before.
Let me step out and set the stage a bit.
This is a question I posted on stackoverflow http://stackoverflow.com/questions/2693 … re-adj-lib One suggestion was archiso. So I started with baseline and was trying to build my 'hello world' iso. Once that is done I will then try port the P.I.N.G scripts to that ISO. Maybe event create it as a package that archiso will then just merged in. Once I get to that point building new iso should be a breeze. Actually, I use the tool as a pxe boot, but until I get the 'hello world' iso working, I can't trust the boot. So I have a script I wrote to build my rootfs once I get the process working and I know it ran once. So I rebooted and all is good.
FYI here is my script.
#!/bin/bash
PINGHOME="/home/Ping"
AIROOTFS="/home/pingbuild/archlive/baseline/work/airootfs"
PINGROOTFS="${PINGHOME}/rootfs"
if [ "$1" == "init" ]
then
rm -r ${PINGROOTFS}
cp -av "${AIROOTFS}/" "${PINGROOTFS}/"
fi
rm -r ${PINGROOTFS}/Braums
cp -av ${PINGHOME}/Braums/ ${PINGROOTFS}/Braums/
dd if=/dev/zero of=initrd bs=1M count=60
mkfs -t ext4 -F initrd
mount -o loop initrd /mnt/PingFS
#rsync -rpltDzv --delete --progress ${PINGROOTFS}/ /mnt/PingFS/
umount /mnt/PingFS
Stepping back in..... When I ran my script I was getting the same errors. Ok, this can't be so I rebooted, ran my script, and it worked. I ran the lsmod and the loop was there.
So tried './build.sh' and it is coping work/airootfs/ to work/mnt/airrootfs.
Not sure why or how it got there since my system is rather new (mid nov 2014) but after a reboot, it does look like systemd 217-8 will insert the loop without forcing a modprobe. That is probably why I did not see it in the wiki.
So back to draining the swamp. If the ISO works and I sure it will then I am back to trying to get pxe boot to work, then integrating the P.I.N.G scripts into the new rootfs. But that maybe another post. :-)