Kristian,
please see inline...
Thanks,
Snjezana
Kristian Rink wrote:
>Hi there;
>
>consequently trying to use the browser on these URLs on the Solaris host
>(via command line and "links" text mode browser):
>
>
>Christopher Kampmeier schrieb:
>
>
>>http://pkg.glassfish.org/v3/dev/solaris-x86/
>>
>>
>
>... is reachable, does show an IPS repository so I think this is okay. Also
>seems to list wxpython, updatetool and friends amongst the packages
>available there.
>
>
>
The trouble is that update center server setup is based on two load
balanced instances, so even if you can access
http://pkg.glassfish.org/v3/dev/solaris-x86/ from your browser there is
a possibility that you are hitting good instance whereas your update
client bootstrap is hitting the bad one. cc-ing Terena - Terena, could
you please check if both instances of this dev repo (and of contrib
repo) are up and running?
>
>
>>origin = http://repine.sfbay:16000/
>>
>>
>
>What kind of host is this? Doesn't looks like it should be publicly
>available, however in my case here is the pkg.glassfish.org URL listed so I
>assume it doesn't matter... :)
>
>
Plase disregard - this is reference to the internal server used for
original package build. Once the initial bootstrap is done, this is
replaced with the URL of the repository specific to your OS. Yes, it
would be nice to clean it up before shipping :-)
>
>
>>prefix = dev.glassfish.org
>>
>>
>
>This host is not available, neither here nor there, but assuming it's not an
>URL anyway I think this also shouldn't be a problem.
>
>
Correct, this is repository prefix used to reference repository and not
an URL.
>
>
>
>>origin-for-sunos-i386 = http://pkg.glassfish.org/v3/contrib/solaris-x86/
>>
>>
>
>contrib access also works flawlessly on the same machine. I don't see any
>problem here.
>
>
Contrib repo is also load balanced, so we have to check if both
repository instances are fine...
>So, nothing I can really add at the moment... also tried different shells on
>the Solaris machine, just to prevent possible flaws in working with zsh or
>bash, but no change. Is there a way to, asides having the 404 spit out,
>determine which URL has been tried to reach when that IOException has been
>thrown? Maybe this would shed some light on this...
>
>Cheers & thanks in advance,
>Kristian
>
>
>