Unable to upgrade VLSR(SRM) "Cannot increment number of connections to <cs p:00007fcd980080b0, TCP:abc.corp.com:80> as max connections (50) already in use"
search cancel

Unable to upgrade VLSR(SRM) "Cannot increment number of connections to <cs p:00007fcd980080b0, TCP:abc.corp.com:80> as max connections (50) already in use"

book

Article ID: 386625

calendar_today

Updated On:

Products

VMware Live Recovery

Issue/Introduction

Unable to upgrade due to timeout issue.
Partially rpm package gets downloaded post which we see max connection timed out messages.

drconfig.log

2025-01-07T14:33:22.694Z verbose drconfig[02095] [SRM@6876 sub=HttpConnectionPool-000000 opID=dd945060-####-458d-8ec5-9177be2####-installUpdate:3d2a] [IncConnectionCount] Cannot increment number of connections to <cs p:00007fcd980080b0, TCP:abc.corp.com:80> as max connections (50) already in use
2025-01-07T14:33:22.694Z verbose drconfig[02073] [SRM@6876 sub=PropertyProvider ctxID=09d570b2 opID=dd945060-####-458d-8ec5-9177be2####-installUpdate:3d2a] RecordOp ASSIGN: info.currentOp, DrConfigConfigurationTask2. Applied change to temp map.
2025-01-07T14:33:22.694Z verbose drconfig[02073] [SRM@6876 sub=PropertyProvider ctxID=09d570b2 opID=dd945060-####-458d-8ec5-9177be2####-installUpdate:3d2a] RecordOp ASSIGN: info.currentOpProgress, DrConfigConfigurationTask2. Applied change to temp map.
2025-01-07T14:33:22.694Z info drconfig[02095] [SRM@6876 sub=Downloading package opID=dd945060-####-458d-8ec5-9177be2e####-installUpdate:3d2a] Downloading: package-pool/open-vm-tools-12.4.5-1.ph4.x86_64.rpm
2025-01-07T14:33:22.694Z verbose drconfig[02095] [SRM@6876 sub=HttpConnectionPool-000000 opID=dd945060-####-458d-8ec5-9177be2e####-installUpdate:3d2a] [IncConnectionCount] Cannot increment number of connections to <cs p:00007fcd980080b0, TCP:abc.corp.com:80> as max connections (50) already in use

 

Environment

VMware Live Recovery 9.X

Cause

The thread pool is consistently consumed because during the HostRescan (which occurs periodically when deviceRecomputations are scheduled by the VC), each ESXi host needs to be processed individually, with each host occupying one active connection.

Therefore, it's recommended to use a larger thread pool when dealing with a large inventory size.

Resolution

  • Please change the timeout values of VLSR.
  • Increase the timeout under the file drconfig.xml
  • Path to file is /opt/vmware/dr/conf/drconfig.xml

    Increase the time out as below

    <defaultClientPoolConnectionsPerServer>200</defaultClientPoolConnectionsPerServer>

         <defaultClientPoolSize>500</defaultClientPoolSize>

  • Save the file and restart SRM services.

Additional Information

Before proceeding with any changes, kindly ensure that a backup of the VLSR nodes is taken.

Powered by Wolken Software