5 Replies Latest reply on Dec 14, 2009 10:53 AM by mrpg

    Agent not checking in, fails to wake up, Unexpected HTTP response code 500

    mrpg

      Folks,

       

      I'm running into an issue with some of my agents-  on the client side all seems fine but they're not checking in or failing to wakeup.  I send the wake-up command- watching the agent monitor I see no wake up activity.

       

      Checking the server task log - subtask details- I see;

      Starting agent wakeup for CLIENT_HOST
      Waking up agent CLIENT_HOST.domain.com using DNS name
      Unexpected HTTP response code 500 received from remote computer

       

      Checking the agent log

      LstnSvr    CAsyncSocket::DoAccept for event: FD_ACCEPT
      LstnSvr    'POST' request received from Host: 10.11.12.13:2588
      SpiPkgr    Error trace:
      SpiPkgr     [Parse SPIPE package of size 366]->
      SpiPkgr      VerifySign error -2147483640
      LstnSvr    Unable to read from the package buffer, GetlastError: 2
      LstnSvr    Bytes received from wakeup call : 366

       

       

      Anyone experience this or/and think/know of a resolution?

       

       

      Thank you.

        • 1. Re: Agent not checking in, fails to wake up, Unexpected HTTP response code 500
          GWIRT

          This looks like the agent is missing the correct server keys or they have gone bad somehow. If you run an "Update Now" on the agent it should try to run the key updater and update the key if a new one has been created.

           

          Also, if you reinstall the agent does that resolve it?

          • 2. Re: Agent not checking in, fails to wake up, Unexpected HTTP response code 500
            mrpg

            Makes sense, this seems to coinside with my FramePkg.exe going bad and giving an error of "sitelist.xml invalid..." if I tried to deploy or install manually.   I was pointed to a KB on stopping services and extracting a back of 2 *key.bin files.   After I completed that process my FramePkg.exe was OK but since that date it seems some clients are in limbo.

             

            I tried doing a bunch of re-deploys, including myself but it didn't seem to work.

             

            *EDIT*  Tried the "Update now" on a few while monitoring the Agent log- still getting;

             

            LstnSvr CAsyncSocket::DoAccept for event: FD_ACCEPT
            LstnSvr 'POST' request received from Host: 10.11.12.13:1263
            SpiPkgr Error trace:
            SpiPkgr  [Parse SPIPE package of size 315]->
            SpiPkgr   VerifySign error -2147483640
            LstnSvr Unable to read from the package buffer, GetlastError: 2
            LstnSvr Bytes received from wakeup call : 315
            LstnSvr CAsyncSocket::DoAccept for event: FD_ACCEPT
            LstnSvr 'POST' request received from Host: 10.11.12.13:1264
            SpiPkgr Error trace:
            SpiPkgr  [Parse SPIPE package of size 315]->
            SpiPkgr   VerifySign error -2147483640
            LstnSvr Unable to read from the package buffer, GetlastError: 2
            LstnSvr Bytes received from wakeup call : 315

             

             

             

            Thank you Greg

             

             

            Message was edited by: mrpg on 12/10/09 8:51:44 PM GMT-05:00
            • 3. Re: Agent not checking in, fails to wake up, Unexpected HTTP response code 500
              mrpg

              So it seems my server keys somehow changed for a few days per gold support???

               

              Now I have, current count, approx 700+ clients with the incorrect key so they're unable to check-in/recieve policies etc.  From the server logs I can see the key hash.  Is there anyway to rectreive the key from the defunct clients and enter it into ePO on the server keys page?  Support is recommending I run frminst /forceuninstall on the defunct clients then redeploy via ePO... I'm going to make that my last option right now- searching fo ran alternative.

              • 4. Re: Agent not checking in, fails to wake up, Unexpected HTTP response code 500
                GWIRT

                Unless you have the sr<servername>.zip to check into ePO, the key's are gone. You can try promoting a new key to master then have all the agents update. The key updater will run and may update the agent keys to the right version.

                • 5. Re: Agent not checking in, fails to wake up, Unexpected HTTP response code 500
                  mrpg

                  I may give support another call, I'm the only one with access to the ePO box and I don't see how the key could change for a few days/week then change back.  I only have one sr*.zip.  Now having to run a forceuninstall on 700+ machines and redeploy is giving me a bit of a headache- if i had total control over all the domains/locations it would be a more mananageable task but I don't and it will be made more complicated that it has to be.

                   

                  Unfortunately I didn't have any backups in place yet to pinpoint a difference from last week- i know, i know- I should have known better.  Now there are 2 SQL dumps, AM and PM to different locations daily.