Jamf Push Proxy return a non-200 status of: 504

I just noticed this in my logs as well, but mine aren't 504s - I'm seeing 503s (several in a row, once per minute), then a 408 or two, then back to the 503s. I am opening a support case to see what this means.

I am now seeing the 504, and I'm noticing that mine show as ERROR versus the WARN in your logs. So far I've only seen it showing up in the logs in the late evening/early morning hours; not sure what to make of that. Jamf Support is asking me if I am utilizing the tasks that the push proxy is used for in the environment - apparently that is:

• Sending Notification Center notifications to computers and mobile devices with Self Service installed
• Requesting management check-ins from personally owned Android devices

We're not doing either of those things, so I'm not sure what is causing this. Stay tuned...

For what it's worth, we get 500 errors when trying to set up a push proxy on our production JSS servers. Something about getting to the Jamf push proxy through a network proxy causes it to fail. That PI is PI-005008. I'm not entirely sure this would be related to that, but if you have network proxies in front of your Jamf boxes that could be part of the issue maybe? Unsure about a resolution though.

Hi all,

Thanks for helping here! submitted an case @jamf and they are digging into it.
They asked me if we are using Android devices because thats maybe have something to do with it

But we do not have android devices in our Jamf Pro environment, so curious what it will be :)

Cheers,
Thijs!

I can't think of any changes recently that would explain this suddenly popping up. I keep an eye on my logs on a regular basis and have never seen these errors before, so I don't know what to think. I tried to renew the push proxy certificate at the request of support and the errors persist.

I'm getting these consistently too

JAMF Push Proxy return a non-200 status of: 504 JAMF Push Proxy return a non-200 status of: 503

We just updated from 9.101 to 10.3.1 and we're seeing these in the (very rarely) 408, 504 and (most often) 503 versions now too, always in connection with the use of InstallApplication with VPP apps in Self Service (not sure why these would be correlated but they seem to be in the logs). Going to open up a ticket of my own.

For those of you that are seeing this - other than these errors, are your logs normal? Mine have been wonky for months; I've had several cases open with jamf due to random errors, and am still battling license monitor runs in the triple digits (sometimes taking up to 8+ minutes). I have [ERROR] [ina-exec-21] [MDMController ] - Error processing mdm request, returning 400. Device: Null, CommandUUID: mraNull in there, feedback tokens for some iPhone (and we don't manage any iPhones...), VPP licenses being transferred to different tokens - all of which I've commented on in other discussions here on jamf nation...

Something odd is going on and based on all of these discussions I know that it's not just me, but I feel like I want to just set fire to my database and start over - and we JUST DID THAT LAST YEAR when we transitioned to cloud hosted (which was to hopefully avoid so many support cases and apparently hasn't helped). Clearly something is wrong, either with this product or our workflows. Wish someone would just finally help me get to the bottom of it, ain't nobody got time for this...

@thejenbot

just like you are mentioning we also battling;
- License monitoring that takes like minutes
- VPP licenses randomly being transferred to other tokens
- Proxy return errors

blah.

For everyone experiencing this issue… have you opened a case with Jamf? Have they provided a PI? The more people report through Jamf Support and attach PIs to their accounts the more traction the issue will get.

oh thank goodness - i'm not crazy then!
jamf is telling me this is normal. i am telling them it is not. the ever increasing amount of errors and junk in my logs i'm being told to just ignore is starting to stack up and take OVER my logs, so much that i have to check them more often because there may be legit things i miss - simply because i now have 500+ lines of pure garbage on a daily basis. this is not how things used to be, and i'm being made to feel like i should just not look at the logs and/or ignore errors when that flies in the face of what i have known to do for years.
blah is absolutely correct.

@emily i do have a case open with jamf but they have not given me a PI. if i do get one i will certainly add it to this case. i've seen several others above have also opened cases with support, hopefully we hear some info soon (other than that this is normal, which i'm not buying).

We have a case open for these errors. We are a MSP that manages 25+ JSSs. The majority of them are at v10.3.0 or 10.3.1, and only one of them has these errors. Our case has been open since 4/12 and we have sent them plenty of logs, etc. Even on the other JSSs that aren't showing these errors in the logs, they seem to take an extremely long time to do VPPLicenseMonitor checks.

Ah alright, we also have already an case open!

Got a response from the fabulous @jordan.hasenberg at jamf: At this point we just need to hang tight until the server performance for jpp.jamfcloud.com is improved - essentially what's happening is your (and every other Jamf Pro) server runs a health check every 60 seconds against jpp.jamfcloud.com, and if jpp returns anything other than HTTP 200 it throws that error in the log. The Jamf Nation team is working to get it sorted out, but in the meantime, there's no tweaking or changes to be made on your end. Most everything should function as expected, other than the intermittent, harmless log chatter that's a side affect of the jpp servers being overloaded. For your notes, it's PI-005741.

I haven't heard back from my Apple SEs yet regarding the long license monitor run times, but this is something jamf is seeing on their end as well, and it's been happening since a VPP outage awhile back. Apparently not related, and most likely something on Apple's end.

Hi there,

Also received an PI number!

We are definitely running into a Product Issue (PI-005741).Like you said, it's a completely harmless error and it at the moment has no workaround. We believe the issue is on our side, and we are working to fix this.I'll attach this case to the Product Issue so that you can trace this back anytime you need.

I received PI-005112 for this, so maybe there are multiple PIs going.