There are a couple different uploaders presently. When some of them upload, it appears the codesize isn't respected. As soon as a couple others do, the big bots get pruned.
Well, I think my uploaders use to be (and probably still are) culprits. But I did not do any tweaking or modification of stock client. I was also under suspicion that even with mine, some of my machines were doing the correct size report and other were not. But they all have identical configs and java distributions.
The only thing which might be of importance: I run everything on OpenJDK, may be it has a different way to evaluate the code size.
Nevertheless, I stopped uploading for a week, so new wrong entries are due to new uploaders. May be we should look for correlation, and see what is the possible reason for the size misreporting.
It would be Skilgannon that runs the rumble server, not I. ;)
No worries =) The rumble is updated, and only accepting contributions from 126.96.36.199
Beaming, it seems your new client is still uploading megabots to the minirumble :-/
Edit: never mind, it now seems to be removing them. Maybe they were left over from the old install?
Edit2: it seems you have two clients running, one is removing and one adding.
Yes, I run 2 clients. More specifically 2 machines, both of which run roborumble and meleerumble, one of this additionally runs meleeTop30. As far as I can see, they have identical relevant software configuration.
I do not get it: microrumble and nanorumble seems to be fine with either of the clients, but minirumble is populated with megabots. Any idea what is going on?
As fat as I see after some tests, it is not a particular client issue. At start up, either of the clients cleans the mess, i.e. removes bogus entries. But then as it runs, it reintroduces them.
Also, both clients complain:
Could not load properties file: ./roborumble/files/codesize1v1.txt
I will report it to Fnl again.
I think I managed to fix the last part of the bug. I made a version 188.8.131.52 here, which is not official (yet). If it works, I will release it pretty much as it is. If not, I will need to work much harder to figure out what causes the problem?!
Now we wait till Skilgannon switches accepted clients in rumble.
184.108.40.206 is now accepted for uploads. I'll disable 220.127.116.11 when it is officially released.
Hi Fnl I've tried the client out a bit, and it seems to correctly remove bots at the beginning of each round, however it doesn't respect codesize for minibots when deciding which rumbles the results of a battle should be uploaded to. This may be due to the priority battles for that rumble requesting those bots, I'm not sure.
Yep, but why is the mciro and nano rumbles are fine? There must be something special about the miniruble.
I will dig into the RoboRumble client sources. I haven't touched RR for a long time, so it is odd that it causes problems with newer versions. Perhaps the change was made around version 18.104.22.168? Anyways. I hope it is okay to upload newer 22.214.171.124 version to check if the problems is fixed when I make new fixes/adjustments to the code? :-)
I am not able to find any obvious bugs on the client side that could cause this trouble. I checked all changes on the RoboRumble client sources since version 126.96.36.199. Could it be some configuration issue or a change on the server side?
Okay. I checked all changed made to the RoboRumble since version 188.8.131.52 and found a potential bug regarding a check for the codesize. I made a "rollback" of the code so it works the same way as for version 184.108.40.206 and earlier. I made a new version 220.127.116.11 that needs to be tested. It looks okay to me when it runs, but I need you to test it thoroughly.
If it works I will make a release of the new 18.104.22.168 ASAP.
It looks like this release fixed the situation. I do not see anything wrong with the minirumble anymore.
Thanks a lot for your effort.
Phew. This one was a hard one to fix. It was almost a one-liner bug, but still hard to find. :-) I will make a official release as soon as possible - hopefully later today. :-)