Hi Team,
I am Prabhjot working on Octopus Deploy. My Jfrog artifactory filenames are too big that when Octopus do deployment the full filename is not present.
Example: abc…file.properties is depolyed as abc…file.prope
Please let me know if there is any limit on filename size.
Thanks,
Prabhjot
Hi Prabhjot,
Thanks for getting in touch!
It could be that you are running into the Windows file path limit (260 characters).
I haven’t seen Octopus automatically shorten a package name before though, would you be able to provide a screenshot of the original package within artifactory and how it is appearing within Octopus and your target?
Regards,
Paul
When you view the deployment task log within Octopus is the file name shortened there too?
I’m wondering if this is just whichever app you’re using in the second screenshot being unable to split the name onto a second line rather than the actual filename being changed.
Hi Paul,
I tried to find in the logs but it just says tar file downloaded but it doesn’t list each and every file inside that.
Please suggest me a way to find out why filenames are getting trimmed.
Thanks,
Prabhjot
Can you send me the deployment task log to review? You can either send me a private message on here with them or email them to support@octopus.com
It does look like the filenames are only being displayed as trimmed when you view them within your terminal window because it limits the characters it can display on each line; I don’t think the actual filename is being changed, though.
Hi Paul,
I tried to find the folders where it download and then extract the files. So the place where it download the package first is in tar format and then extract it to new folder where the file names are trimmed.
One thing I checked is that when I untar the package from where it downloaded first the filenames are not trimmed. So maybe the command used to extract files is where something going wrong.
I also tried to create a file with too long name and it is showing fully without trim.
Can you please let me know which command is being used to extract the files
Thanks,
Prabhjot
We use the sharpcompress library to handle extraction.
I’ve run a test of this on Windows and Linux and I do not see any trimming of file names, but it could be specific to the OS you’re deploying to.
I don’t see any specific issues raised around long file names being trimmed with sharp compress, though.
If it is a limitation of the extraction library then you may need to reduce the length of the file names within your package.
Ok, what step is being used to perform the copy?
Or is this a custom script being used?
As mentioned, if you can download and send over the full task log we can investigate this further.
Hi @prabhjotkour.91,
Thanks for sending over the log file.
Would you be able to check the initial extract folder in order to find out if the file names appear correctly there?
Afterwards, if you’re able to check the custom extract folder to compare any differences you see.
We’ve run tests internally on a tar package with long names and so far we haven’t been able to observe any errors occurring but that’s not to say it isn’t possible.
Kind Regards,
Adam
Hey @prabhjotkour.91,
Thanks for sending that over and for testing that.
It’s strange that the file names are being changed so much, would you be able to send over screenshots with a simple LS command in the directory to remove the permission/timestamp information on the left, please?
Additionally, would it be possible for you to send over the tar file for us to test ourselves, to see if we can replicate the issue? You can upload it to our support files repository here: Support Files.
Kind Regards,
Adam
Hi Adam,
I did just ran ls command and the output is the same.
Thanks,
Prabhjot
Hi @prabhjotkour.91,
Thanks for confirming, would you be able to share the tar package with us?
I provided a link to upload in my last response: Support - Octopus Deploy
Is this something you’re able to send over? It would help with our testing to be able to have a 1:1 replication.
Kind Regards,
Adam