Home Help Center


mwilsonmwilson Member Mark WilsonOrganization: Integrated Computer Solutions, Inc.Project: TBD ✭✭✭

Our locator service (from version 3.0.0, apps now using 3.1.0) failed last night. The service had been up for several months prior to that, but with a pretty light load.

Just wondering if you guys have any data on MTBF for the locator under heavier loads than we have subjected it to here (our load will increase a lot very soon)?



  • marcmarc FloridaAdministrators, ZeroC Staff Marc LaukienOrganization: ZeroC, Inc.Project: The Internet Communications Engine ZeroC Staff
    No, we don't have any such data, nor do I believe that it can be measured. If we run a long-term test of IceGrid (or any other Ice component), and it fails, we consider such failure a bug, and eliminate this bug. Thus, if we run the same long-term test again, it will not fail anymore due to this bug, rendering the previously measured time interval to this failure meaningless.

    Of course, there might be operating system failures, hardware failures, etc. However, none of these failures have anything to do with Ice in particular, but depend on your computing environment.
  • benoitbenoit Rennes, FranceAdministrators, ZeroC Staff Benoit FoucherOrganization: ZeroC, Inc.Project: Ice ZeroC Staff
    Hi Mark,

    Do you have any information on the failure such as a core file with a stack trace? Are you using replica groups? There's a known bug with replica groups which could result in a crash of the registry, see [thread=2745]here[/thread] .

Sign In or Register to comment.