![]() |
Resuming partial downloads How do you get Limewire to resume a partial download? I have not been successful, but can do it all the time in Bearshare. |
The answer is simple: You can't. I have tried it for months now, and it has never worked. Nobody in this forum knows how to do it either, as that topic is usually posted about three to four times a week and noone has ever got a helpful response from any user or moderator. If you need to resume your partial downloads you should think about migrating to Phex which is also cross-platform - but it doesn't look as nice as LimeWire ;-) . Whatever Lime Group claims about resuming 'across hosts', i am sure that noone has ever been able to resume anything with Lime. And I am also sure that none of the moderators will dare posting a reply. Don't get me wrong; except for the nonexistant resuming feature I am a great fan of Lime and I do advise all my friends to use it. But I don't like the way Lime Group claims that LimeWire has features it just doesn't have, and I especially mislike the way of remaining silent in the shadows whenever this question is posted. I have also written an email to the LimeWire support but i only got excuses about too many requests that couldn't be answered. |
I'm really not sure how this rumor got started; LimeWire has always supported resumed downloads! Here's how it works: If you are downloading a file, and the server disconnects you, LimeWire will automatically try to resume the file. If the resumes don't work after several attempts, it enters the "Couldn't Download" state. If you hit the "Force Resume", LimeWire will try the resume process all over. If you kill LimeWire while the download is in progress, LimeWire will try to resume the download next time it runs. Now, if you have several hosts in a download group, LimeWire will even try to resume the file from another host in the group. If that file couldn't be downloaded, LimeWire will try other files (e.g., with different sizes or names) in the group. However, the original download attempt stays in the Incomplete directory for your convenience. Here's what LimeWire can't do. First, LimeWire does not provide a way to re-search for a file and add that file to an existing download group. It may do so in the future. Secondly, it doesn't support resumptive push uploads. That's not really a problem, since the push route will have expired by the time you get to resume, but we'll implement it in an upcoming version. Regards, Christopher Rohrs Senior Software Engineer LimeWire |
RE: Resuming partial downloads I follow most of your explanation, and I have not tried the Force Resume when it says, Couldn't Download. However, I have tried and tried to get a file that has, in a previous session, downloaded say 90% of the file size. On a subsequent use of the program, after closing and reopening Limewire, and searching for and finding the exact named file with the exact same file size,and trying to download it again, after verifying that a 90% partial exists in the incomplete directory, Limewire starts at the beginning again. Bearshare will ALWAYS start at the end of the partial. I cannot justify my time using Limewire any further until this problem is fixed. Kencam |
Right, that's what I mentioned in the last paragraph above; although LimeWire will resume in every way possible from the hosts of a smart download group, you can't add new search results to a group. We will try to add this feature in a future release. In the meanwhile there is a dirty hack you can use to resume from different searches. If you look at your Incomplete files, you will see that they are of the form "N(slot)-(size)-(name)" for normal downloads, where (slot) is a number that starts at 0 and is always increasing, (size) is the file's size, and (name) is the file's name. (Push downloads have a different form; let's not worry about that.) This scheme allows you to download two copies of the same file at the same time without interleaving them on disk, e.g., in case you want to try two hosts in parallel. By renaming these files appropriately, you can force new downloads to resume from old search results. You can even force LimeWire to resume from a file with a different name. Again, this will probably be automated in the future. Thanks for the feedback. Christopher Rohrs Sr. Software Engineer LimeWire |
I have already tried it that way with LimeWire 1.3 and 1.4, and the only effect was that the existing part was deleted and the download began at 0% again. Furthermore, I know well how you expect the Resume Feature to work, but I can tell you that it has never worked on my machine. Normally, a transfer gets interrupted because the user on the other end killed the upload or disconnected from the net. a resume try will therefor have no effect at all. I have never observed that LimeWire actually was able to connect to another servent of that group and resume the file, even if there were 15 entries in the group. And was I wrote was true: This question is posted over and over and I am not the first one who complained about not getting answers to it. (just look for old threads and you will see) I have asked many limers an how it is done, and everyone told me that they never got a download resumed properly, just as i didn't., that's why i don't think that it is a fault of my system. |
Resumes Thanks for the tips. I will try the rename with a new search to see if I can make that work. I appreciate your answers, apparently others are much more frustrated. You sound like you are involved in the development of Limewire. I think you have some great features, and with a solid resume feature, you would knock Bearshare off the shelf. I have noticed, though, that the speeds I get with Bearshare are much higher than with Limewire. I have Roadrunner, so I have a much larger range of possibilities than many others. On Bearshare, I see frequently downloads of 220 to 280 kbps by others from my computer, and I can download from others frequently at over 300 kbps. Huge files, over 100 megs take less than one hour. On Limewire, however, I have never seen a transfer on my machine going either direction ever exceed about 60 or 70 kbps. Thanks again. Kencam |
Resuming Partials By the way, would you be so kind as to describe what "renaming appropriately" means? Thanks, Kent |
Sure. Say you originally tried to download a file named "a.txt" that was 1000 bytes long. Then you will see a temporary file named Nx-1000-a.txt in your Incomplete directory, where x is some seemingly random slot number. (Again, we're assuming normal downloads only. Push downloads are a slightly different beast.) Now, say you want to resume where the download left off from a file named b.txt that is 999 bytes long. You should rename the incomplete file to Ny-999-b.txt, where y is LimeWire's next "slot number". Figuring out y isn't so easy. If you have just started LimeWire and have no downloads, y should be 0. It is incremented after every subsequent download. In any case, when you try to download "b.txt" from your search result window, LimeWire will be fooled into trying the resume from the incomplete file for a. Yup, it's a pain and we're planning on improving this. Keep in mind, however, that you probably won't have to do this if you really take advantage of LimeWire's smart grouping feature; go for groups with four stars and lots of entries, and LimeWire's "resume across hosts" feature will do the work for you. -Christopher Rohrs Sr. Software Engineer LimeWire |
It worked for me... Well, the Resume feature works for me. If I select the file for which the "Download Failed" and click the resume button, it picks up the download from where it left off. Caveats: I've done this within minutes after I lost my dial-up connection, so the same file is most likely available from the same source. I haven't tried it hours or days later. |
Your feature is coming... LimeWire 1.5 will support your "resume from old download groups" feature. -Christopher Rohrs Sr. Software Engineer LimeWire |
well, i dont mean to bitch but when you download a 675 megs file and your download stops at 500 megs, it aint fun to hear that you actually cant resume the download... especially when you are on a 56k modem like I am... please fix this quickly because otherwise you'll have a lot of users that will use other gnutella-based programs... |
I have the same complaint about resuming downloads. Often times when i connect i am trying to get a file that is quite large, and and have only a few hosts to choose from when i do smart download. So i'm pretty much screwed if the download gets killed or interrupted. "Screwed" meaning starting that 200meg download all over again. i'm going to try Phex... |
LimeWire 1.5 will do exactly what you want, I think. We've built the installers and are testing in-house now. Should be available to the public next Monday or Wednesday. I think you'll like it. -Christopher Rohrs Sr. Software Engineer LimeWire |
ok, i'll look for it. i don't really want to leave Limewires great interface, and Phex requires that i install 32 megs of java software to run correctly. So if Limewire can deliver on resuming broken downloads, i'll stay loyal! |
ok, i'll look for it. i don't really want to leave Limewire's great interface, and Phex requires that i install 32 megs of java software to run correctly. So if Limewire can deliver on resuming broken downloads, i'll stay loyal! |
Phex doesn't need any more Java Software than LimeWire. Just install the Java Runtime Environment (JRE, 7 MB), not the entire JDK. The JDK is for writing Java programs, the JRE is all you need for using them. PS: What exactly will LimeWire 1.5 do ? Research for a file ? Handle a dynamic list of download sources ? Download the way Phex or Gnucleus do ? PPS: If LimeWire's grouping feature is for resuming Smart Downloads, why does it actually group by name and not by size ? you cannot resume from a file that has a different size! |
I saw someone mention that most "transfer interrupted"s were caused by the users. I cannot believe that. I have watched close to 99% of uploads afflicted by the bug. Hard to believe that many computers have connection problems. I've looked around this site, and this is the closest I can come to finding any mention of it. If I'm doing something wrong, please help me. Running 1.6 on a Mac with 9.1 and 512 MB of RAM. Everything else runs great. I like the LimeWire, and would like to let users upload, but I'm gonna **** a lot of people off if they can't get through a download. I actually watched an ip address get 1% at a time, get interrupted, force resumption, get another 1%, and so on until he got to 100. Of course there were times in between when all the upload slots were taken, so he/she is out of luck for the moment. Then they get through, and somebody else gets bumped off. Looks like a bandwidth problem on the Limey side to me. Please tell me I've got a setting incorrect so I can fix this thing and make it work.!!!! |
All times are GMT -7. The time now is 11:50 AM. |
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2025, vBulletin Solutions, Inc.
SEO by vBSEO 3.6.0 ©2011, Crawlability, Inc.
Copyright © 2020 Gnutella Forums.
All Rights Reserved.