Hi John, Yes, Java monitor reporting zero instances running would rule this out completely. Have you set WOHost on wotaskd, JavaMonitor, and the instances? It could possibly be a timing issue leading to one of them (probably wotaskd) using a different WOHost name.
Chuck On 2014-08-09, 6:42 AM, "John Pollard" wrote: Hi Chuck, Would the fact that JavaMonitor reports 0 instances running rule that possibility out? There are instances running, but it would seem wotaskd doesn't know it, and as such that is the state shown in JavaMonitor. Otherwise my adding a delay before restarting wotaskd would have made the problem worse. However, I will certainly try restarting apache the next time it happens. I just suspect that the netstat error may be related as it is similarly intermittent and wotaskd would need to find out similar information that netstat returns, presumably. Thanks John On 8 Aug 2014, at 19:54, Chuck Hill <ch...@global-village.net<mailto:ch...@global-village.net>> wrote: Hi John, Another thing to try when this happens is to restart Apache. I have seen cases where something happens and it caches a state of no instances. My guess is that may be it starts too early before wotaskd is ready. After a restart the instances become visible again. If that results in no change, at least you can rule this out. Chuck On 2014-08-08, 1:18 AM, "John Pollard" wrote: Hi list, Despite adding in delays when starting wotaskd on reboot, the WO services have again (intermittently) failed to start processes smoothly on reboot, which I believe is wotaskd starting processes but then not being able to confirm they are running on the relevant sockets, so trying to start them again. We have a script that monitors CLOSE_WAIT issues using netstat -a, and I note the log file intermittently shows "getnameinfo failed" from the netstat -a command. Could this be the same issue that causes wotaskd to intermittently fail on startup? I can find very little on the web about this failure message or how to resolve, other than that is may be a DNS resolution issue, but why intermittent? The server is running on Amazon Linux. Multiple domains route to the same server static IP address We don't specifically assign a hostname to the server, instead we access via ssh on user@domain Amazon Linux creates a hostname, which currently reports as: ip-10-73-155-123 but this has never featured in any of our configuration and is not the same IP as the server's public static IP address that all our domains point to. Any thoughts on whether this getnameinfo failed might relate to the WO startup issues? The failure only happens occasionally it seems, but does seem to happen in clumps. Thanks John _______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-deploy mailing list (Webobjects-deploy@lists.apple.com<mailto:Webobjects-deploy@lists.apple.com>) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/webobjects-deploy/chill%40global-village.net This email sent to ch...@global-village.net<mailto:ch...@global-village.net> _______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-deploy mailing list (Webobjects-deploy@lists.apple.com<mailto:Webobjects-deploy@lists.apple.com>) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/webobjects-deploy/jpollard%40inrax.com This email sent to jpoll...@inrax.com<mailto:jpoll...@inrax.com>
_______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-deploy mailing list (Webobjects-deploy@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/webobjects-deploy/archive%40mail-archive.com This email sent to arch...@mail-archive.com
