Topics created by jstarbird_7831

Jon, I haven't found too much on this besides what we already recommended. I did find this option specifically for Import-module : https://stackoverflow.com/a/28512164

I did try that and it works outside Docker where it just uses the user that is running it. Within Docker though I can't get it to work. I tried giving it creds or api key and it just won't validate. On proget it never shows an attempt when it's the api key either. Anyway, this work around is working so now i'm progressing. I'll pass on the details once I have it working. Thanks.

Eric, thank you! That was the issue, I didn't know about that issue.

Keep in mind that Timeout (or cancellation in general) is not guaranteed, and requires that the task (generally, an operation) properly check for a cancellation token. For example, an operation that simply did a Thread.Sleep for 1000 seconds could not be timed out after 10 seconds. In a case of a stuck execution, restarting the service is the best way to clear things. So, we'd need more context to identify if it's a bug in the operation, or simply something that isn't support.

Just to close this post: this was fixed in InedoCore 1.0.12, after diagnosing the issue on a ticket (EDO-5683).

This is also fixed (BM-3271) starting in v6.1.0

Thanks, it is working now.

Ok, thanks. I understand what you're saying now.

Alana, thanks. I didn't realize those variables did not appear in that list, thought that was all variables. In any event, knowing they are app settings variables helps. Thanks!

I was finally able to go through the source code to see that the menu option was in fact missing in v1.0.3 - I don't know how I missed that you were on that version vs. the one I was using (v1.0.4) :( An upgrade would solve, but I also see your ticket and will prod a developer to take a look at that for you as well and get out a new hub installer with the improved debug logging.

It autocomplete/dropdown uses Username & Password credentials for the Package Source URL. I also agree, that it should support both or at least be clear which type it's using :) The tracking issue for this is here: HH-17

Good idea about the throw statement - I've added a feature request with suggested formal grammar to our public issue tracker as IEE-14. There are other things we'll have to take into account such as documentation updates and updating the visual plan editor to handle this, so it's not scheduled yet, but at least it is logged.

I found a solution. I had to change the pipeline to not restrict the order of the stages.

I believe the issue is related to the "Execution in Progress Page", which is the "live view" page of the execution, and shows logs like a big textbox. The slowness seems to be more related to client interaction, at least in our exploration. So we have decided to limit this view to previous ~500 entries in v6; you will notice a big improvement for such large log, and can see full details in the details page.

Ben, I didn't set it for any particular stage. Based on what you're saying though I think the issue was that didn't have the Listener in place when the app had completed the previous stage. I assumed the email would only go out when the Stage the approval was attached to was initiated for deployment but it sounds like it would do so when the previous stage completes, is that correct?

I'm not really a RegEx expert, but what you're describing should work. Without seeing the input, it's difficult to say. I use http://regexstorm.net/tester to test it every now and then, since we use the .NET RegEx engine, so it might be helpful to make a test case in there first. FYI: you can see exactly how the code uses regex, it's quite simple: https://github.com/Inedo/inedox-inedocore/blob/master/InedoCore/Common/Operations/Files/ReplaceFileTextOperation.cs Can you share something that appears to work on RegExStorm , but not in the OtterScript plan? And then share the OtterScript plan?

That's what I was afraid of. Thanks.

Alex, thanks. I don't want to update it via DB I am just going to be changing one and wanted to verify everyplace it's used before I made the change.

Apparently this "works" if you do: IIS::Ensure-AppPool Test ( Runtime: "", Enable32BitAppOnWin64: true, Pipeline: Integrated ); Might be better to do something like: set $NoManagedCode = ""; IIS::Ensure-AppPool Test ( Runtime: $NoManagedCode, Enable32BitAppOnWin64: true, Pipeline: Integrated );

Jon I've run into the same issue before dealing with the npm feeds. I ended up hosting the proget site twice in IIS under different hostnames. In IIS we have two websites: proget.mycompany.com Windows Auth enabled Forms Auth disabled Anonymous Auth disabled formsauth-proget.mycompany.com Anonymous Auth enabled Both sites use the same physical path of your proget installation. The host name in the binding must be different though. That is so IIS knows which way to go, the website with Windows auth or the site with the anonymous auth. We've been using this for several months with no problems. I hope it helps in your situation. Please note, I do not work for ProGet, I'm a fellow customer offering assistance.