Squelch

Hey Guys, My last post may have come over a bit terse. I was annoyed at having to fault find what I thought was a router problem, but was in fact Apex playing up. I always welcome new features and improvements, and in the past have suggested them. These new features and improvements need to be tested. After all that is what beta code is. The thing that concerned me the most was the fact that the new version had quickly been established as being unstable, but it was still the most predominant version on the download page two weeks later. Problems do happen and need to be fixed. Downgrading ver 1 to alpha, or flagging the possible problems on the news page would have saved me some time and hassle, and I hazard a guess, other users too. I hope you can fix the problem soon, and without too much hair pulling. I look forward to trying it out. I volunteer myself for testing anything you want to try out. Just don't scare off new customers with dodgy code on the front page :)

How is this progressing? I'm just about to set up a new Tuxbox and would like to help if I can. I'm not a programmer, but do have passing knowledge of C++, networks and shell scripts, and have good fault finding skills with an eye for user interfaces. Any specific areas that need attention or research I'm happy to do. So a progress update with particukar problems and needs that need attention would be useful. [edit] Nevermind. I just managed to open the other thread http://forums.apexdc.net/index.php?showtopic=1065 I kept getting time outs for some reason and assumed it was a dead link. [/edit]

Is this anywhere near resolution, or is the advice still to roll back to the previous version? I just recently had cause to fire up Apex and downloaded the latest version 1.0.0 b2(auto update didn't see new ver) This was yesterday 23/04 I now have the same crash errors as other users here (exception.txt ommited for brevity) So the question is, why is a very unstable version still on the download page since the 10/04? I submit that a version jump from 0.4.0 to 1.0.0 would indicate a feature complete and major bug free version. Any new users to Apex might be very dissapointed, and therefore pass on the belief that it is very unstable. Does this issue only affect upgraded clients, or is it the same for new installs?

Hi, I've recently been warned on one of my favorite hubs that multisource clients affect the hub in a bad way. As I'm not entirely sure in what way, and a search of various sources hasn't turned up anything, could someone please enlighten me. Is this just an automatic reaction, or is there some foundation to it?

Hi darko79, I have the same problem and didn't spot that a username might be the problem. Are you able to open the filelists entry in the queue? Apex crashes as soon as I mouse over the entry. Tbh, the entry is redundant, because the queued filelists are shown in the progress window anyhow.

Hi all, Is there a log analyser that can be used with ApexDC++? I've tried DC++ Log Analyser, Clones DCStatGenerator, and dC++Lv. None of them seem to work with Apex logs. If there isn't such a beast, it might be worth a feature request or addon. Being able to identify for sure that a user that is always downloading, but never has an open slot for you would be very useful.

Hehe, which ones? I'll back them up so I don't lose the info, and see if the newly generated ones are useable.

That would explain it then. A misunderstanding in what the log would contain. I thought it might be the formatted transfer history for that user. Perhaps a change of option description to "Open User Message Log" or something might help. Thanks for clearing that up. Ps I can't edit topic title. It's not a bug so could someone change and move it to [support] instead.

Hi guys, I've recently discovered the "Open user log" right click command in favorite users. Even though logs for that user are present, I receive "User log not exist". Is this a bug, or specific to my installation?

Log analyser seems to freeze when processing the logs. It eventually gives up with an exception, and never gets past uploads.log. It's version 0.992 I might have corrupt logs. I can't view user logs from within Apex either. If anyone can confirm it works with Apex logs then I'll look into it further. I might have to resort to drastic measures and reinstall Apex. :thumbsup:

I'm not sure what you mean there. I have all logging turned on except personal and main chat. My config has the correct path too "\ApexDC++\Logs\" I just can't view them. I'm tempted to re-install Apex again but I have a big queue and don't want to lose all of the other data as well. :thumbsup:

I think the hash table is simply that. The file list references the index for that file in the table. I'm not sure if adding files different to the default tree would cause problems, but I'm only after a subset of the default tree so the files have been hashed anyhow. I had trouble with topic tool button after forum update too. Try a hard page reload (Ctrl-F5)

Ah! So it doesn't work out as I thought then. But surely if another user is queueing for that file, they get to connect on my disconnection. Does the slot remain opened? Is the pause user defineable?

Isn't the "Auto grant extra slot" option for favorite users a partial method of what bigcanuck is asking for? If this could be implemented hub wide on the enabled hub in favorites, wouldn't this achieve what is required regardless of multiple user id's on different hubs? If a truncated version of the default filelist is presented to a hub, then surely searches would still give the correct resposes to another user. If they are on another hub, and see the other content, then they could only download via your id on that hub. It appears to me, that the same user (even with different nick) on different hubs, is treated as different users even though they have the same content. Therefore changing the filelist presented as that user doesn't matter. Both of these seem related.

An update, Using the Automatically add downloaded files to share option has a big impact on performance. It appears that hashing the new file increases CPU usage greatly, and the disk io causes the freezing. Turning this option off improves the situation, but doesn't cure it completely.

Hi all, I have two problems, which may be related. 1. Tab window keeps going blank combined with a 5 to 20 second pause in the client. Downloads appear to still take place, but cpu useage is very high during pause. This occurs shortly after connecting to a hub in a zombie state. ie main window shows connected message then stays frozen until a manual reconnect is done. (all other connected hubs are ok) This problem is tolerable at first, but the whole client then starts to slow down completely over an extended period (several hours), often resulting in me closing it and restarting. I have only noticed this since upgrading to 0.3.0. 2. Own file list does not display. I cannot view my own file list. The tab appears briefly, but then disappears. Is it possible to delete and recreate my list? Any help would be appreciated. [edit] I have noticed another minor bug in the gui. When manualy connecting from favorites window, the main window enters the restore state, whereas it was maximised before. Not a real killer but annoying all the same. [/edit]

I've noticed that using automatically add downloads to share does increase CPU usage. Hashing the new file does cause an apparent freeze. Turning this option off helped me a bit, but hasn't cured it totally.

I'm not sure if this is also causing what I have observed. On opening Apex, the memory used is about 64Mb. After a 14 hour session it has grown to 124Mb. Watching the performance graph in procces explorer shows that there is a steady increase in the private memory used. I have some screen shots if they help.

So it appears ALL multisource clients are banned. To be honest, and judging by the language used, the ops on the hub are a lost cause. The opinion I got, was that they suspect ApexDC++ as being a leechers tool. Looking through my logs, and from observation, the up/down flow is about even for the users on that hub. It seems they were not interested in even entertaining the thought it might even be. I'm of the opinion, that being truthful about your action pays dividends. My description says "Using ApexDC++ with no limits" Some hub ops I've found to be very open to it, and a couple admitted to using PeerWeb, and contemplating a change. Even though some hub scripts fail to recognise Apex, and emulation is the only way to get in. It's only right to not appear underhand. I used the emulate function when first switching to Apex as I was already registered on many of the hubs while using DC++. The hub that banned me found out I was using a multisource client somehow. Maybe because I had advertised it elswhere, or through some othe means. Apart from reconnecting using DC++ under another nick to finish a couple of incomplete files, I'm not tempted to return to that particular hub. I did send pm's, signing my original nickname, to my favorite users there, explaining that I'd been banned, and the reason. I also told them about TGO so they could continue downloading from me. Perhaps they can carry the message back. Am I right in thinking that segmented downloads are fairer? I've noticed that large files come in smaller blocks, even if there is only one source. After finishing a segment, there is an idle pause before starting the next, and sometimes I get a "No slots available" message and have to wait. Surely this is better than holding a slot for a very long time, and gives other users a chance with the possiblity of getting the missing portions from them.

Hi, I couldn't find anything like this mentioned elswhere. Is it possible to add a virus scan command option for finished downloads? Something along the lines of; "\path to command line scan prog\virusscan.exe %1" with full path to file being passed as a parameter. Hope this isn't a daft request.

It's a great idea, Going a little further, are profiles a possibility? Having 2 or 3 different share trees for different hubs. Some hubs don't like music and films, so having a subset of the default share only for that hub would be nice.

I'm sorry to say I finally got banned from the hub. I had tried to strike up dialogue with one of the ops, but they failed to see any benefit in ApexDC++. Ignorance is bliss I suppose. Doesn't DC++ emulation inhibit segmented downloading from that particular hub? If so what complaint does the hub still have? I've made no secret of the fact I use ApexDC++, in fact it's in my default description now. I've only enabled emulation to get around the "No Tags" script. Do you have a FAQ outlining the pro's and cons? It might be useful to have a "one stop" hub owners and OP's info pack. BTW As there were some good sharers on the hub in question. I pm'd my favorites and pointed them at TGO so they can continue to share with me.

Thank you for the info guys, I'm not using any torrent software, and the only other things using the line are browser and mail client. I do auto connect to around 20 hubs, so that must be what causes it. Thanks for the tip on closing and reopening. It does work, especially on two of the most troublesome hubs that I now know to share a server. I've turned my limiter on to a modest setting, only to be flamed by some hubs for doing it. It's a shame people don't realise ApexDC++ is not the leechers client they see it as. Oh well - spread the word

Hi all, I keep getting the message "*** An existing connection was forcibly closed by the remote host." usualy after about a minute of starting a new session, and successfuly connecting to registered hubs. Sometimes it settles down after a few reconnects, but other times it persists. What causes this? Am I being rejected from the hubs because of some misconfiguration or problem my end?

Thanks for that. I couldn't find any information on what it meant.