PDI offline

[gersonfreir]

Again, for the second time las 5 days, my PDI (Personal Development Instance) is offline, and even after I attempt to start it and click the refresh button, it remains offline. When I try to access it via the URL, a "502 Bad Gateway snow_adc" error appears

[Kh_Samyak · Accepted solution]

Hi @gersonfreir,

Thanks for reporting this, and sorry for the trouble. We checked with the PDI team. New software installations across the PDI infrastructure have temporarily taken several PDIs offline, which is likely why you’re seeing the 502 error. The affected instances will remain unavailable until the installations are complete.

Thanks for your patience in the meantime.

[VJ_Srivastava · Accepted solution]

Hello @gersonfreir,

From an infrastructure architecture perspective, frequent 502 Bad Gateway errors on a PDI are a direct consequence of operating on shared, ephemeral hardware.

ServiceNow packs thousands of free PDIs onto shared physical nodes. If a "noisy neighbor" on your specific node executes a massive runaway script or triggers a memory leak, it can crash the underlying container. When that happens, the Application Delivery Controller (snow_adc) loses its connection to your specific instance and throws the 502 error.

This inherent instability is exactly why enterprise development pipelines mandate decoupling code from infrastructure. You must treat your PDI as entirely disposable hardware. If you are regularly committing your scoped applications to a remote Git repository via the ServiceNow Studio Source Control integration, hardware degradation becomes a non-issue.

You can simply release the failing infrastructure, provision a new instance, and pull your complete codebase down in minutes rather than fighting unstable hardware.

[Kh_Samyak · Accepted solution]

Hi @gersonfreir,

Thanks for reporting this, and sorry for the trouble. We checked with the PDI team. New software installations across the PDI infrastructure have temporarily taken several PDIs offline, which is likely why you’re seeing the 502 error. The affected instances will remain unavailable until the installations are complete.

Thanks for your patience in the meantime.

[23a3sa42d2 · Accepted solution]

Hello Kh_Samyak,

Thank you for checking with the PDI team and explaining the issue.

My PDI, dev295422, is still unavailable. Could you please let me know if there is any update on the infrastructure installations or an estimated timeframe for restoring access to the affected instances?

I appreciate your help. Thank you.

[VJ_Srivastava · Accepted solution]

Hello @gersonfreir,

From an infrastructure architecture perspective, frequent 502 Bad Gateway errors on a PDI are a direct consequence of operating on shared, ephemeral hardware.

ServiceNow packs thousands of free PDIs onto shared physical nodes. If a "noisy neighbor" on your specific node executes a massive runaway script or triggers a memory leak, it can crash the underlying container. When that happens, the Application Delivery Controller (snow_adc) loses its connection to your specific instance and throws the 502 error.

This inherent instability is exactly why enterprise development pipelines mandate decoupling code from infrastructure. You must treat your PDI as entirely disposable hardware. If you are regularly committing your scoped applications to a remote Git repository via the ServiceNow Studio Source Control integration, hardware degradation becomes a non-issue.

You can simply release the failing infrastructure, provision a new instance, and pull your complete codebase down in minutes rather than fighting unstable hardware.