Problem with cron jobs

[SteveEngelking]

Im using cron-job.org default version and it throws an error:

Failed (output too large)

This apparently means that the application is throwing back to much data and that is limited to 64KB. Is there anyway to resolve this apart from me having to buy an expensive subscription? Or am I doing something wrong?

Steve

[Jefferson49]

For the export, I am using the same code like webtrees. That means the creation of the export is different, but the export procedure is the same. Therefore, I think there should not be any limitation in the order of 64kB. The procedure might reach some limits for large mega bytes.

I cross checked on my own server. My daily GEDCOM export (saved to the webtrees data folder) is about 5MB. That works even on a NAS with less performance.

What comes to my mind:

• Can you explain what kind of activity you trigger with the cron-job and which kind of export shall be performed?
• Can you explain in more detail, why the data might be limited to 64kB? Do you get an error message? Can you post it?
• Maybe, the export really creates too much data. For example, if it hangs up in a loop and contiues to write the same data forever. Can you try to run the same export in the browser front end and see what happens? Can you access the (large) data even if the error occurs and check the end of the file?
• Is it possible that the module creates a PHP error ? I could imagine that there might be a limit for error texts. Do you have a possibility to read the output file (even if it is too large)?

[SteveEngelking]

Thanks for your reply. When I run the export in webtrees manually it works fine. This is the command in cron-job.de https://www.rothenhof.org/index.php?route=/ExtendedImportExport&tree=tree5&key=sack080725!? it creates this error: Cronjob execution: Failed (output too large) https://www.rothenhof.org/index.php?route=/ExtendedImportExport&tree=tree5&key=sack080725!? It brings the same error when I do a test run. Maybe there is something wrong with my command line? Steve

…

On 5. Sep 2025, at 18:25, Jefferson49 ***@***.***> wrote: Jefferson49 left a comment (Jefferson49/ExtendedImportExport#37) <#37 (comment)> For the export, I am using the same code like webtrees. That means the creation of the export is different, but the export procedure is the same. Therefore, I think there should not be any limitation in the order of 64kB. The procedure might reach some limits for large mega bytes. I cross checked on my own server. My daily GEDCOM export (saved to the webtrees data folder) is about 5MB. That works even on a NAS with less performance. What comes to my mind: Can you explain what kind of activity you trigger with the cron-job and which kind of export shall be performed? Can you explain in more detail, why the data might be limited to 64kB? Do you get an error message? Can you post it? Maybe, the export really creates too much data. For example, if it hangs up in a loop and contiues to write the same data forever. Can you try to run the same export in the browser front end and see what happens? Can you access the (large) data even if the error occurs and check the end of the file? Is it possible that the module creates a PHP error ? I could imagine that there might be a limit for error texts. Do you have a possibility to read the output file (even if it is too large)? — Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BLDVB3F4E3NYGMNQBWH2ZR33RG2OVAVCNFSM6AAAAACFQOKLMKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENJYHE2TANBYGA>. You are receiving this because you authored the thread.

[Jefferson49]

The URL looks different in my installation:
/webtrees/index.php?route=webtrees/ExtendedImportExport

It seems that "webtrees" is missing in your route, but maybe this is because my installation runs on "http://my_site/webtrees/"

How about the "!?" at the end of the URL? At least "?" is a reserved character in URLs and should not be used in passwords. Or included within "", e.g. key="mypassword?"

However, I still cannot see your error message. Can you paste it in plain text? I did not understand the link above.

Please delete your message above, if it contains the real password key!

[SteveEngelking]

Changed to a plain password If I do a test run it and change &tree=tree5 to &tree= i.e. https://www.rothenhof.org/index.php?route=/ExtendedImportExport&tree=&key=sack080725 It runs fine but does not export the file to the server.

…

On 5. Sep 2025, at 19:47, Jefferson49 ***@***.***> wrote: Jefferson49 left a comment (Jefferson49/ExtendedImportExport#37) <#37 (comment)> The URL looks different in my installation: /webtrees/index.php?route=webtrees/ExtendedImportExport It seems that "webtrees" is missing in your route, but maybe this is because my installation runs on "http://my_site/webtrees/" How about the "!?" at the end of the URL? At least "?" is a reserved character in URLs and should not be used in passwords. Or included within "", e.g. key="mypassword?" However, I still cannot see your error message. Can you paste it in plain text? I did not understand the link above. Please delete your message above, if it contains the real password key! — Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BLDVB3DZDX3GAF4H4QRT3VL3RHEC7AVCNFSM6AAAAACFQOKLMKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENJZGI3DAMBUGY>. You are receiving this because you authored the thread.

[SteveEngelking]

See attached test run result

…

On 5. Sep 2025, at 19:47, Jefferson49 ***@***.***> wrote: Jefferson49 left a comment (Jefferson49/ExtendedImportExport#37) <#37 (comment)> The URL looks different in my installation: /webtrees/index.php?route=webtrees/ExtendedImportExport It seems that "webtrees" is missing in your route, but maybe this is because my installation runs on "http://my_site/webtrees/" How about the "!?" at the end of the URL? At least "?" is a reserved character in URLs and should not be used in passwords. Or included within "", e.g. key="mypassword?" However, I still cannot see your error message. Can you paste it in plain text? I did not understand the link above. Please delete your message above, if it contains the real password key! — Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BLDVB3DZDX3GAF4H4QRT3VL3RHEC7AVCNFSM6AAAAACFQOKLMKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENJZGI3DAMBUGY>. You are receiving this because you authored the thread.

[SteveEngelking]

Did another test with a tree with only 1 record and it ran without error but did not save it to the server. So it definitely has something to do with the size of the gedcom file. But that does not explain why it isn’t saved.

…

On 5. Sep 2025, at 20:18, Steve Engelking ***@***.***> wrote: <Screenshot 2025-09-05 at 20.17.42.png>See attached test run result > On 5. Sep 2025, at 19:47, Jefferson49 ***@***.***> wrote: > > > Jefferson49 > left a comment > (Jefferson49/ExtendedImportExport#37) > <#37 (comment)> > The URL looks different in my installation: > /webtrees/index.php?route=webtrees/ExtendedImportExport > > It seems that "webtrees" is missing in your route, but maybe this is because my installation runs on "http://my_site/webtrees/" > > How about the "!?" at the end of the URL? At least "?" is a reserved character in URLs and should not be used in passwords. Or included within "", e.g. key="mypassword?" > > However, I still cannot see your error message. Can you paste it in plain text? I did not understand the link above. > > Please delete your message above, if it contains the real password key! > > — > Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BLDVB3DZDX3GAF4H4QRT3VL3RHEC7AVCNFSM6AAAAACFQOKLMKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENJZGI3DAMBUGY>. > You are receiving this because you authored the thread. >

[Jefferson49]

To save the GEDCOM file on the server, you need to add "&action=save" or "&action=both" to the URL. In case of "both", it will download and save in parallel. If no action is specified, the file will be downloaded as default.

You can also define default settings in the control panel. If no value for "&action=" is provided, the default value is used. However, I do not recommend this for scripts, because it is better practice to provide all parameters in the script and the the script shall be independent of any default settings.

[Jefferson49]

You should not use " &tree=". This will request for a tree with an empty name and should result in an error "Tree not found".

[Jefferson49]

A good approach for testing is to write test URLs in a text editor like notepad++. Afterwards click on the URL in the text editor or copy the text to the browser command line. With this approach, you can open and test the URL with your browser. If the intended action is performed correctly, you can use the same URL in a script afterwards.

One good thing about testing with the browser is that you get error messages in the browser, e.g.

This helps to identify errors in the URL etc.

[Jefferson49]

See attached test run result

Seemingly, the attachement is not shown in the GitHub issue. Maybe, this does not work if answering with an email.

[SteveEngelking]

All seems to be working fine now. Thanks so much for your great support.
Would it be an idea to add the options in the settings and generate the complete URL? (It already knows what tree you re in anyway and the save command is needed anyway) Might help idiots like me!