Current HEAD state

Here you can post threads specific to the current release of the core (099)

Moderator: POL Developer

Post Reply
xeon
Forum Regular
Posts: 338
Joined: Fri Oct 31, 2008 3:18 am
Location: Italy

Current HEAD state

Post by xeon »

I was thinking about upgrading my POL099 to the latest build.
What is the current state of the HEAD? Is it stable, production ready? Should I wait for something?
Any comment from who's using one of the latest build?
OWHorus
Grandmaster Poster
Posts: 105
Joined: Sat Feb 04, 2006 1:24 pm
Location: Vienna, Austria

Re: Current HEAD state

Post by OWHorus »

Hello,

please wait some days - I am trying this. We have a rather big script system, not the official distro, and use a lot of functions. So if the newest version works for us, it will probably work for you too. I cannot say anything about the official distro.

I am using Debian wheezy. We use an older core (0.99, from around 2012), this ran for over two years, but it seems to be unstable under certain conditions, which I just found out. We had only one server crash before, for unknown reasons. I currently use a 32-bit POL core compiled under Debian, statically linked.

So - I will report here about our results.

OWHorus
bodom
Former Developer
Posts: 140
Joined: Sat Feb 21, 2015 7:52 pm
Location: Italy

Re: Current HEAD state

Post by bodom »

The are rumors of a new release coming out soon (finally).

It should be a 2nd Beta or even a Release Candidate :grouphug:
xeon
Forum Regular
Posts: 338
Joined: Fri Oct 31, 2008 3:18 am
Location: Italy

Re: Current HEAD state

Post by xeon »

Thanks OwHorus, I'll wait your comments

We have strange crashes during shutdown, but we've always had these, both with 099 and older cores. Figure it out...
OWHorus
Grandmaster Poster
Posts: 105
Joined: Sat Feb 04, 2006 1:24 pm
Location: Vienna, Austria

Re: Current HEAD state

Post by OWHorus »

Hi xeon,

there seem to be problems with the current HEAD. One of the problems I want to analyze further, but we have a situation where POL send masses of packages and hangs the client. But this could be a problem only with our client (5.06a) and our setup.

The other problem is uoconvert - see my bug report in an extra post.

The crash during shutdown has happened to me one or two times - no idea what it is, but it always saved correctly. This was - as far as I remember - never on Linux, but on the Windows version, but the same server version (POL.099 from around 2012). We had a Windows Server before 2013 for the game, and there I always shut down with a command from inside the game.

OWHorus
OWHorus
Grandmaster Poster
Posts: 105
Joined: Sat Feb 04, 2006 1:24 pm
Location: Vienna, Austria

Re: Current HEAD state

Post by OWHorus »

OWHorus wrote:Hi xeon,

there seem to be problems with the current HEAD. One of the problems I want to analyze further, but we have a situation where POL send masses of packages and hangs the client. But this could be a problem only with our client (5.06a) and our setup.

The other problem is uoconvert - see my bug report in an extra post.

The crash during shutdown has happened to me one or two times - no idea what it is, but it always saved correctly. This was - as far as I remember - never on Linux, but on the Windows version, but the same server version (POL.099 from around 2012). We had a Windows Server before 2013 for the game, and there I always shut down with a command from inside the game.

OWHorus
Hello,

solved the client crash mentioned above: It was one of our older scripts, which should send packets in certain situations, and did send a malformed packet for multis due to a core change. (I am not trying to apportion blame here, I just forgot that this script existed, so it was my fault)

To make it clear for all (corechanges.txt mentions it, but not very clearly):

The script API for Multis has changed, uoconvert generates a different multis.cfg (Graphic entry missing), but also the member multi.graphic is now always 1 if you read it via script! The correct new way to generate multi.graphic if you need it in scripts is now:

graphic := multiref.multiid + 0x4000;

if you use a client with a maxtileid = 0x3fff.

The core seems to run stable now, I did not see any problems with it. Uoconvert is fixed now, I tested it. I will further test, because we have a lot of special mechanisms (we do not use the distro). But for now it does look good.

OWHorus
xeon
Forum Regular
Posts: 338
Joined: Fri Oct 31, 2008 3:18 am
Location: Italy

Re: Current HEAD state

Post by xeon »

Thanks OWH for the feedback.
I'm thinking upgrading as we have just incurred in a few core bug from the revision we're using, like a splitstackinto() bug when done to a stack of item in a storage area.
Post Reply