[23:20 GMT] We’re back!
[23:15 GMT] Running the startup script again.
[23:04 GMT] Tested successfully, bundling again…
[22:49 GMT] Found the error: init-ajaxlife.sh: line 1: mono: command not found
[22:18 GMT] Startup script failed – it terminated the instance less than a minute after starting it. Trying to work out why…
[22:08 GMT] Watching the startup script run…
[21:44 GMT] Waiting for the bundling to finish.
[21:17 GMT] Image conversion should work now. Testing in progress…
[20:56 GMT] Image conversion is a pain. Still trying. >.>
[19:58 GMT] Running assorted tests
[19:41 GMT] Configuring AjaxLife server
[19:09 GMT] Rebuilding from scratch. 0% complete.
[UPDATE] As predicted, it’s down. Attempts to repair it are being foiled by such annoyances as lessons and wireless connections that drop constantly.
Due to accidental deletion of the manifest.xml file that goes with AjaxLife’s EC2 AMI, there is a reasonable chance that it’s going to break tomorrow and not be able to restart automatically.
I will work on this as soon as I get the chance; I’m going to sleep now (yawn).
Sorry to hear that kath, depends on the viewer i suppose, and the machine, and how the lindens are feeling today, and how the particular server is feeling today, etc.
Chilko, this isn’t LL’s server. It’s amazon’s server. And I accidentally deleted the file (courtesy of someone mislabelling the “delete” button as a “view” button…)
Yes, i know, i was referring to your microblog post on TSLP.
Hi, Katharine!
Greetings from the OpenSim and OSGrid team. I wanted to let you know that the latest release of AjaxLife crashes OpenSim regions when attempting login. I’ve reported it as a bug on our bugtracker. If you want to look into it on your end, maybe you could determine where the bits are going wrong. I don’t know if it would help you at all, but you can find the error report here: http://opensimulator.org/mantis/view.php?id=491
Thanks and keep up the great work,
paulie Femto (OpenSim bug squisher)
ps: I apologize for spelling your name three different ways (all incorrect) in the error report.
If only the mislabeled, invisible ‘recover file’ button were easier to find/implement.
It would probably get some good usage too..
Fear my necroposting powers! >.>
You can’t necropost on a blog. Blog posts never move out of chronological order.
paulie, apologies for the very late response – Akismet got excited and blocked your comment (having blocked absolutely nothing for about a week, leaving me to be spammed with moderation emails) without telling me.
Anyway, it looks like I might be able to hack it into working by disabling the agent appearance request, which both looks to be the place it’s crashing and is sent immediately after receiving the MOTD, and is thus a fairly likely culprit. This’ll have the disadvantage of the client remaining ruthed, however, unless moving the request elsewhere would work.
This appears not to be the case – I disabled the de-ruthing and it still crashed anyway.
Thanks for looking into it. No apologies necessary for the late response. I’ll let our team know you’re onto it.