build.sh fails at STEP 14/18 [SOLVED]

[shark711]

When trying to build with podman using build.sh, i get the following error:

STEP 14/18: RUN cd /tmp && unzip DaVinci_Resolve_Linux.zip && if [ "od -An -j 8 -N 6 -x --endian=big *DaVinci_Resolve_*_Linux.run" = " 4149 0200 0000" ]; then sed '0,/AI\x02/{s|AI\x02|\x00\x00\x00|}' -i DaVinci_Resolve_Linux.run; fi && ./*DaVinci_ResolveLinux.run --appimage-extract && cd squashfs-root && ./AppRun -i -a -y && cat /tmp/squashfs-root/docs/License.txt && rm -rf /tmp/.run /tmp/squashfs-root /tmp/.zip /tmp/*.pdf
Archive: DaVinci_Resolve_Linux.zip
inflating: DaVinci_Resolve_18.6.6_Linux.run bad CRC 77fc7613 (should be ecc631bd)
inflating: Linux_Installation_Instructions.pdf
Error: building at STEP "RUN cd /tmp && unzip *DaVinci_Resolve_Linux.zip && if [ "od -An -j 8 -N 6 -x --endian=big *DaVinci_Resolve_*_Linux.run" = " 4149 0200 0000" ]; then sed '0,/AI\x02/{s|AI\x02|\x00\x00\x00|}' -i *DaVinci_ResolveLinux.run; fi && ./*DaVinci_Resolve_Linux.run --appimage-extract && cd squashfs-root && ./AppRun -i -a -y && cat /tmp/squashfs-root/docs/License.txt && rm -rf /tmp/.run /tmp/squashfs-root /tmp/.zip /tmp/*.pdf": while running runtime: exit status 2

Not sure if the "bad CRC" is causing the exit code (2) but i redownloaded the file, and also tried with latest v19 and get the bad CRC each time

$ podman images
REPOSITORY TAG IMAGE ID CREATED SIZE
docker.io/library/rockylinux 8.6 8cf70153e062 2 years ago 202 MB

$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 24.04 LTS
Release: 24.04
Codename: noble

$ glxinfo | grep "OpenGL version"
OpenGL version string: 4.6.0 NVIDIA 535.183.01

$ nvidia-smi
Tue Jul 16 20:42:27 2024
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.183.01 Driver Version: 535.183.01 CUDA Version: 12.2 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA GeForce RTX 3080 Off | 00000000:2B:00.0 On | N/A |
| 0% 37C P8 22W / 320W | 467MiB / 10240MiB | 6% Default |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| 0 N/A N/A 2790 G /usr/lib/xorg/Xorg 206MiB |
| 0 N/A N/A 3072 G /usr/bin/gnome-shell 61MiB |
| 0 N/A N/A 4715 G ...,262144 --variations-seed-version=1 104MiB |
| 0 N/A N/A 6860 G /usr/bin/gnome-text-editor 10MiB |
| 0 N/A N/A 11977 G /snap/snap-store/1124/bin/snap-store 24MiB |
+---------------------------------------------------------------------------------------+

$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 43 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 32
On-line CPU(s) list: 0-31
Vendor ID: AuthenticAMD
Model name: AMD Ryzen 9 3950X 16-Core Processor
CPU family: 23
Model: 113
Thread(s) per core: 2
Core(s) per socket: 16
Socket(s): 1
Stepping: 0
Frequency boost: enabled
CPU(s) scaling MHz: 48%
CPU max MHz: 4761.2300
CPU min MHz: 2200.0000
BogoMIPS: 7000.17

Keen to hear if others had this issue.
Recently got rid of Windows and switch to linux, and unable to get this up and running

[fat-tire]

The line:

inflating:bad CRC 77fc7613 (should be ecc631bd)

Suggests that there is actually a corrupted .zip file. CRC is a cyclic redundancy check, used to ensure a file does not contain any errors. If a .zip fails a CRC check while uncompressing, it usually means something in the file is not what was expected.

If you like, you can use unzip -t <filename>.zip to verify the CRC outside of the build script (on the host side) to see if the file can in fact be uncompressed successfully.

I know you said you tried re-downloading the .zip, but are you 100% sure you replaced the older file with the newly downloaded one? (And did you notice any difference, say in file size in the two downloads?)

You can easily find if the downloaded files are identical or not:

Method 1:

$ diff file1 file2
Binary files file1 and file2 differ

(if they are the same, you will not get any output)

Method 2:

$ sha256sum file1
$ sha256sum file2

(each of the above commands will output a random-looking characters --a "hash" of the file. If the characters outputted are the same for each file, the file is the same. Otherwise, they differ.)

Also, instead of 18.6.6, maybe you want to try a different file-- say the latest beta for version 19, just to be 100% sure it's not that file itself. (Maybe it is corrupted on Blackmagic's side?)

Let me know if any of these ideas helped!

[shark711]

Deleted all my downloaded zip files

Then
$ cd ~/containers/resolve
$ wget https://swr.cloud.blackmagicdesign.com/DaVinciResolve/v19.0b5/DaVinci_Resolve_19.0b5_Linux.zip?verify=[REDACTED] -O DaVinci_Resolve_19.0b5_Linux.zip

Checked for potential files with same name to be sure:
$ sudo find / -iname "*davinci*.zip"
/home/shark711/containers/resolve/DaVinci_Resolve_19.0b5_Linux.zip
find: ‘/run/user/1000/doc’: Permission denied
find: ‘/run/user/1000/gvfs’: Permission denied

$unzip -t /home/shark711/containers/resolve/DaVinci_Resolve_19.0b5_Linux.zip
Archive: /home/shark711/containers/resolve/DaVinci_Resolve_19.0b5_Linux.zip
testing: DaVinci_Resolve_19.0b5_Linux.run OK
testing: Linux_Installation_Instructions.pdf OK
No errors detected in compressed data of /home/shark711/containers/resolve/DaVinci_Resolve_19.0b5_Linux.zip.

Deleted all my podman images podman rmi -f $(podman images -q) (only had one, but wanted to make sure)

ran build.sh again
Got similar error (different version this time, so different file)
STEP 14/18: RUN cd /tmp && unzip DaVinci_Resolve_Linux.zip && if [ "od -An -j 8 -N 6 -x --endian=big *DaVinci_Resolve_*_Linux.run" = " 4149 0200 0000" ]; then sed '0,/AI\x02/{s|AI\x02|\x00\x00\x00|}' -i DaVinci_Resolve_Linux.run; fi && ./*DaVinci_ResolveLinux.run --appimage-extract && cd squashfs-root && ./AppRun -i -a -y && cat /tmp/squashfs-root/docs/License.txt && rm -rf /tmp/.run /tmp/squashfs-root /tmp/.zip /tmp/*.pdf
Archive: DaVinci_Resolve_Linux.zip
inflating: DaVinci_Resolve_19.0b5_Linux.run bad CRC 79b765c8 (should be 336201b6)
inflating: Linux_Installation_Instructions.pdf
Error: building at STEP "RUN cd /tmp && unzip *DaVinci_Resolve_Linux.zip && if [ "od -An -j 8 -N 6 -x --endian=big *DaVinci_Resolve_*_Linux.run" = " 4149 0200 0000" ]; then sed '0,/AI\x02/{s|AI\x02|\x00\x00\x00|}' -i *DaVinci_ResolveLinux.run; fi && ./*DaVinci_Resolve_Linux.run --appimage-extract && cd squashfs-root && ./AppRun -i -a -y && cat /tmp/squashfs-root/docs/License.txt && rm -rf /tmp/.run /tmp/squashfs-root /tmp/.zip /tmp/*.pdf": while running runtime: exit status 2

Funny thing is, when I ran it again after deleting the podman images, I got a different CRC number (without touching the zip file)
inflating: DaVinci_Resolve_19.0b5_Linux.run bad CRC 81538d94 (should be 336201b6)
vs
inflating: DaVinci_Resolve_19.0b5_Linux.run bad CRC 79b765c8 (should be 336201b6)

checked original file to be safe, and was still ok
~/containers/resolve$ unzip -t DaVinci_Resolve_19.0b5_Linux.zip
Archive: DaVinci_Resolve_19.0b5_Linux.zip
testing: DaVinci_Resolve_19.0b5_Linux.run OK
testing: Linux_Installation_Instructions.pdf OK
No errors detected in compressed data of DaVinci_Resolve_19.0b5_Linux.zip.

Going to wipe my hard-drive and reinstall Ubuntu from scratch

[shark711]

sigh.....
I give up

Complete!
warning: /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm: Header V3 RSA/SHA256 Signature, key ID 8483c65d: NOKEY
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 325M 100 325M 0 0 48.9M 0 0:00:06 0:00:06 --:--:-- 52.7M
Verifying archive integrity... Error in check sums 3600907162 3558731085
Error: building at STEP "RUN if [[ dnf list libxcrypt-compat == libxcrypt-compat ]]; then export EXTRA_PACKS="${EXTRA_PACKS} libxcrypt-compat" ; fi && if [[ dnf list compat-openssl11 == compat-openssl11 ]]; then export EXTRA_PACKS="${EXTRA_PACKS} compat-openssl11" ; fi && if [[ dnf list compat-openssl10 == compat-openssl10 ]]; then export EXTRA_PACKS="${EXTRA_PACKS} compat-openssl10"; fi && export NVIDIA_VERSION=$NVIDIA_VERSION && export ARCH=$ARCH && dnf update --refresh -y && dnf install dnf-plugins-core xorg-x11-server-Xorg libXcursor unzip alsa-lib librsvg2 libGLU sudo module-init-tools libgomp xcb-util python39 -y && dnf install epel-release -y && dnf install xcb-util-cursor -y && if [ ! -z "${EXTRA_PACKS}" ]; then dnf install ${EXTRA_PACKS} -y ; fi && if [[ "${NO_PIPEWIRE}" == 0 ]] ; then dnf install libXi libXt libXtst procps dbus-x11 libSM pipewire libcurl-devel apr apr-util libXinerama libxkbcommon libxkbcommon-x11 libXrandr xcb-util-image xcb-util-keysyms xcb-util-renderutil xcb-util-wm libglvnd-opengl pulseaudio-libs nss libXcomposite libXdamage -y && curl https://vault.centos.org/8-stream/AppStream/x86_64/os/Packages/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm -o /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm && dnf -y remove pipewire-alsa && rpm -i --nodeps --replacefiles /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm && PINNEDSHA=/usr/bin/sha256sum /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm && if [ "${PINNEDSHA}" != "a870db3bceeeba7f96a9f04265b8c8359629f0bb3066e68464e399d88001ae52 /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm" ]; then echo "bad checksum" ; exit 1; fi && rm /tmp/alsa-plugins-pulseaudio-1.1.9-1.el8.x86_64.rpm ; else dnf install alsa-plugins-pulseaudio -y ; fi && curl https://us.download.nvidia.com/XFree86/Linux-${ARCH}/${NVIDIA_VERSION}/NVIDIA-Linux-${ARCH}-${NVIDIA_VERSION}.run -o /tmp/NVIDIA-Linux-${ARCH}-${NVIDIA_VERSION}.run && bash /tmp/NVIDIA-Linux-${ARCH}-${NVIDIA_VERSION}.run --no-kernel-module --no-kernel-module-source --run-nvidia-xconfig --no-backup --no-questions --accept-license --ui=none && rm -f /tmp/NVIDIA-Linux-${ARCH}-${NVIDIA_VERSION}.run && rm -rf /var/cache/yum/* && dnf remove -y epel-release dnf-plugins-core libcurl-devel && dnf clean all": while running runtime: exit status 2
Build of resolve:19.0b5 is complete. To run resolve, try:

if I run it again, i am back to same error
STEP 14/18: RUN cd /tmp && unzip DaVinci_Resolve_Linux.zip && if [ "od -An -j 8 -N 6 -x --endian=big *DaVinci_Resolve_*_Linux.run" = " 4149 0200 0000" ]; then sed '0,/AI\x02/{s|AI\x02|\x00\x00\x00|}' -i DaVinci_Resolve_Linux.run; fi && ./*DaVinci_Resolve_Linux.run --appimage-extract && cd squashfs-root && ./AppRun -i -a -y && cat /tmp/squashfs-root/docs/License.txt && rm -rf /tmp/.run /tmp/squashfs-root /tmp/.zip /tmp/*.pdf
Archive: DaVinci_Resolve_Linux.zip
inflating: DaVinci_Resolve_19.0b5_Linux.run bad CRC 78bb12c5 (should be 336201b6)
inflating: Linux_Installation_Instructions.pdf

Keeping in mind this is a fresh install of ubuntu, so there is nothing installed other than the prerequisites listed

[shark711]

I have good news and bad news
Good news, after running it a 3rd time, it passed that step
Bad news is, it failed on step 18/18

STEP 18/18: CMD __NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia __VK_LAYER_NV_optimus=NVIDIA_only /opt/resolve/bin/resolve
COMMIT resolve:19.0b5
Getting image source signatures
Copying blob 879054335f94 skipped: already exists
Copying blob bc70bd2f43a5 done |
Error: committing container for step {Env:[ARCH=x86_64 ZIPNAME=DaVinci_Resolve_19.0b5_Linux.zip EXTRA_PACKS= POWERTOOLS=powertools NVIDIA_VERSION=535.183.01 BUILD_X264_ENCODER_PLUGIN=0 USER=resolve USER_ID=1000 NO_PIPEWIRE=0 PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin] Command:cmd Args:[__NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia __VK_LAYER_NV_optimus=NVIDIA_only /opt/resolve/bin/resolve] Flags:[] Attrs:map[] Message:CMD __NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia __VK_LAYER_NV_optimus=NVIDIA_only /opt/resolve/bin/resolve Original:CMD __NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia __VK_LAYER_NV_optimus=NVIDIA_only /opt/resolve/bin/resolve}: copying layers and metadata for container "aacd196888932bf3e68bd4edbff206b959ac1ab3669775f14b271d34987454d3": writing blob: storing blob to file "/var/tmp/container_images_storage37436088/1": happened during read: Digest did not match, expected sha256:bc70bd2f43a50b2feccabf7cf7bb5740c8469b4a2bdd954e181ff54dc44ecd26, got sha256:b3542456d20ef625daa9e7bc555157e6bd0fcb9566b7144020cd7b237119f04f

... can only assume my Samsung SSD 970 EVO Plus 2TB has issues, cos i don't understand what else this could be

[fat-tire]

Yes unfortunately with the inconsistency of these errors, and the fact that they seem to be related to checksums and verifying data, it looks like you may be having SSD issues :( At the very least, it may be a good idea to back up your most important data ASAP, then maybe replace the drive. Also, you might check the dmesg command (requires sudo on recent versions of Ubuntu) and look for any i/o errors or other warnings...

[shark711]

Yeah we are going beyond the realm of this container and i appreciate all the help you have provided @fat-tire

I don't expect you to solve this, but for what it is worth
All the disk checks came back with good results, including sudo fsck -cc /dev/nvme0n1p1, sudo badblocks -v /dev/nvme0n1, smartctl (long and short tests). Took a while for them all to complete

$ sudo badblocks -v /dev/nvme0n1
Checking blocks 0 to 1953514583
Checking for bad blocks (read-only test): done
Pass completed, 0 bad blocks found. (0/0/0 errors)

$ sudo smartctl -a /dev/nvme0n1
...
Num Test_Description Status Power_on_Hours Failing_LBA NSID Seg SCT Code
0 Short Completed without error 11944 - - - - -
1 Extended Completed without error 11944 - - - - -

the ONLY thing i found was
$ sudo smartctl -a /dev/nvme0n1
...
Error Information (NVMe Log 0x01, 16 of 64 entries)
Num ErrCount SQId CmdId Status PELoc LBA NSID VS Message
0 5411 0 0x0008 0x4004 - 0 0 - Invalid Field in Command
...

And this doesn't appear to be a disk issue, not sure if this is perhaps firmware/driver issue

[fat-tire]

That's good that you're not seeing too many errors... I don't know what that SMART error means actually. SMART was popular with hdd, but I don't see it referenced so much with ssd lately. Did you keep an eye on dmesg when you did experience the checksum errors? (You can follow it "live" with sudo dmesg -w) Often if you see i/o errors there you'll know something is wrong.

Also, did you try a fsck -fy /dev/nvme0n1 as well as the -cc option? cc wouldn't make any changes or attempt at repairs, while fy would... but if you're not seeing any errors to fix maybe it wouldn't make a difference.

Anyway, hope you finally got the container working :)

[shark711]

I found the problem. One of my memory modules are bad

https://x.com/sh4rk711/status/1814226197586522269

Explains the inconsistencies

your recommendation on dmesg started to list segmentation faults on random commands, such as apt-get update etc. which made me look at memory, thanks for that @fat-tire

[shark711]

Yip that was it... thanks for your help @fat-tire

Complains about it not supporting CUDA 12.3, but after manually selecting openCL and my GPU it starts up

$ ./resolve.sh nvidia-smi
non-network local connections being added to access control list

No "container-machine-id" file found for this container.
Creating new one derived from host machine-id (94c36052f6d346f091bf819b7aaaaf08)...
This container's machine-id file can now be found at /home/shark711/containers/resolve/mounts/container-machine-id and may be replaced if needed.
The container's /etc/machine-id : 8f3d6233f82cff65b7643e94392c8225
The network driver setting is : --network=none
Bind-mounted directories of interest :
CONTAINER (CentOS 8) -> HOST (Ubuntu)

/home/resolve -> /home/shark711/containers/resolve/mounts/resolve-home
/opt/resolve/logs -> /home/shark711/containers/resolve/mounts/logs
/opt/resolve/configs -> /home/shark711/containers/resolve/mounts/configs
/opt/resolve/easyDCP -> /home/shark711/containers/resolve/mounts/easyDCP
/opt/resolve/.license -> /home/shark711/containers/resolve/mounts/license
/opt/resolve/'Resolve Disk Database' -> /home/shark711/containers/resolve/mounts/database
/var/BlackmagicDesign/DaVinci Resolve -> /home/shark711/containers/resolve/mounts/BlackmagicDesign
/opt/resolve/Media -> /home/shark711/containers/resolve/mounts/Media
Fri Jul 19 21:09:38 2024
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.183.01 Driver Version: 535.183.01 CUDA Version: 12.2 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA GeForce RTX 3080 Off | 00000000:2B:00.0 On | N/A |
| 0% 36C P8 19W / 320W | 489MiB / 10240MiB | 0% Default |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
+---------------------------------------------------------------------------------------+

[fat-tire]

Terrific! I mean, not that you had bad RAM, but that you figured the issue!