Select to view content in your preferred language

Web Adaptor + IIS woes on Azure IaaS Virtual Machine

82
0
a month ago
MuneerMajid
Occasional Contributor

I am currently troubleshooting a weird issue with a single machine 11.3 deployment on Azure VM which has a Portal, server and Data store installed. FME Flow is also installed on this same VM alongside Flex LM.

The Web Adaptors for Portal and Server are also installed on the same VM with /portal and /server context. The site is open to the public domain with the DNS allowed via Azure Front Doors and WAF rules. 

We have been experiencing issues with the site becoming non responsive quite frequently and the services not being reachable via the Web Adaptor, and then the services URL becomes responsive on its own after a few minutes. The same happens via the Portal Web Adaptor URL as well. 

MuneerMajid_0-1737133618217.png

There are times when I have found that the IIS App Pools were stopped and I have updated their recycling properties for them to refresh once every night. There are other times when the site doesnt respond over :7443 and :6443 as well. 

Investigative Steps performed so far - 

1) I have confirmed the VM resources and ensured the CPU, RAM and network performance are not taxed. We even added RAM initially and I can confirm resources are good. 

2) Reinstalled Web Adaptors again which didnt solve the issue

3) Uninstalled IIS completely and reinstalled it, mounted a new SSD disk and reconfigured IIS Home wwwroot into this drive, created a new site with new bindings, reinstalled web adaptors here. Issue still persists. 

4) Looked through all IIS Logs for the last several days, dont see any issues there

5) Have looked to Portal and Server logs, dont see any errors relating to this particular issue

6) Added a second AGS server on a different VM and federated with the portal, kept the DNS and web adaptor of the second server on its on VM instead of installing it on the primary one. Moved essential services to this new server via python and recreated maps and apps with the new services, so far so good, but the that only isolated any probable issues with the Server. The portal still becomes unresponsive occasionally. 

Last resort, I am considering deploying a fresh multimachine enterprise site and migrating some of the existing content over. Meanwhile, I am still looking for a probable cause for this issue. No idea whats causing it. Any recommendations or suggestions will be deeply appreciated. 

 

 

0 Kudos
0 Replies