Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
On Sun, Dec 27, 2009 at 10:34 AM, SmallSister development wrote: > Evan Daniel wrote: >> Persistent-temp is not part of the datastore, and is not expected to >> stay within the datastore size limit. It's used to hold data from >> partially completed downloads. In theory, it will only grow to be >> slightly larger than your total queue size. However, there appear to >> be some bugs that make it leak space slowly. If it grows beyond your >> total queue size rapidly, though, that's not something I've seen. > > I can live with some additional disk-space use for "downloaded files in > transit"; if it cleans up after the downloads are complete. (I planned a > few gigs for downloads I configured Freenet.) Yet, presistent-temp grew > over the weeks to achieve 11 GB... with several wrapper-induced > shutdowns of Freenet. (The db4o implementation for downloads can get > pretty unresponsive at times.) > > But with 17GB datastore (15 GB configured) PLUS 11GB in persistent-temp > Freenet eats the best part of a 30 GB partition... Reclaiming ~9GB in > unused temp files would make quite a difference. > The workaround for persistent-temp is to shut down the node, then delete node.db4o and persistent-temp-*/ and restart the node. Warning, this will remove all in-progress uploads/downloads, so let those complete first. Evan Daniel ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
Evan Daniel wrote: > Persistent-temp is not part of the datastore, and is not expected to > stay within the datastore size limit. It's used to hold data from > partially completed downloads. In theory, it will only grow to be > slightly larger than your total queue size. However, there appear to > be some bugs that make it leak space slowly. If it grows beyond your > total queue size rapidly, though, that's not something I've seen. I can live with some additional disk-space use for "downloaded files in transit"; if it cleans up after the downloads are complete. (I planned a few gigs for downloads I configured Freenet.) Yet, presistent-temp grew over the weeks to achieve 11 GB... with several wrapper-induced shutdowns of Freenet. (The db4o implementation for downloads can get pretty unresponsive at times.) But with 17GB datastore (15 GB configured) PLUS 11GB in persistent-temp Freenet eats the best part of a 30 GB partition... Reclaiming ~9GB in unused temp files would make quite a difference. ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
On Fri, Dec 25, 2009 at 7:18 AM, SmallSister development wrote: > Dennis Nezic wrote: >> On Thu, 24 Dec 2009 23:03:25 +0100, SmallSister development wrote: >>> Dennis Nezic wrote: On Thu, 24 Dec 2009 13:28:36 +0100, SmallSister development wrote: > 2) Error message on the file sharing pages: > Internal error > Internal error > Return to queue page. > [The "queue page" gives this error message too] I have posted about this bug as far back as June 12th '09, with freenet version 1216 and probably earlier -- basically ever since the db40 thing was merged :P. Toad flippantly suggested it was my "flaky hardware" (raid 5 software array) or a bug in the db40 libraries. But the regularity of these errors (I still get them from time to time, and have to delete my node.db4o) makes me surmise otherwise. >>> The error disappeared when I freed some disk space... And I got the >>> data store shrunk by resizing it; it is at 40% use now. I'll clean >>> out temp files and node.db4o once the downloads have completed. >> >> Interesting. Come to think of it, I also run pretty low on free >> diskspace ... 99% disk usage (of over 1tb) not being uncommon ... >> although that should still leave several gigs free. But, it's also >> interesting that you didn't have to delete your node.db4o file to get >> rid of the "Internal error". > > And the "Internal error" reappeared after Freenet filled up the > partition again overnight. "persistent-temp" had grown to 11 GB, which > is a significant chunk of a 30 GB partition. Persistent-temp is not part of the datastore, and is not expected to stay within the datastore size limit. It's used to hold data from partially completed downloads. In theory, it will only grow to be slightly larger than your total queue size. However, there appear to be some bugs that make it leak space slowly. If it grows beyond your total queue size rapidly, though, that's not something I've seen. Evan Daniel ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
Dennis Nezic wrote: > On Thu, 24 Dec 2009 23:03:25 +0100, SmallSister development wrote: >> Dennis Nezic wrote: >>> On Thu, 24 Dec 2009 13:28:36 +0100, SmallSister development wrote: 2) Error message on the file sharing pages: Internal error Internal error Return to queue page. [The "queue page" gives this error message too] >>> I have posted about this bug as far back as June 12th '09, with >>> freenet version 1216 and probably earlier -- basically ever since >>> the db40 thing was merged :P. Toad flippantly suggested it was my >>> "flaky hardware" (raid 5 software array) or a bug in the db40 >>> libraries. But the regularity of these errors (I still get them >>> from time to time, and have to delete my node.db4o) makes me >>> surmise otherwise. >> The error disappeared when I freed some disk space... And I got the >> data store shrunk by resizing it; it is at 40% use now. I'll clean >> out temp files and node.db4o once the downloads have completed. > > Interesting. Come to think of it, I also run pretty low on free > diskspace ... 99% disk usage (of over 1tb) not being uncommon ... > although that should still leave several gigs free. But, it's also > interesting that you didn't have to delete your node.db4o file to get > rid of the "Internal error". And the "Internal error" reappeared after Freenet filled up the partition again overnight. "persistent-temp" had grown to 11 GB, which is a significant chunk of a 30 GB partition. ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
On Thu, 24 Dec 2009 23:03:25 +0100, SmallSister development wrote: > Dennis Nezic wrote: > > On Thu, 24 Dec 2009 13:28:36 +0100, SmallSister development wrote: > >> 2) Error message on the file sharing pages: > >> Internal error > >> Internal error > >> Return to queue page. > >> [The "queue page" gives this error message too] > > > > I have posted about this bug as far back as June 12th '09, with > > freenet version 1216 and probably earlier -- basically ever since > > the db40 thing was merged :P. Toad flippantly suggested it was my > > "flaky hardware" (raid 5 software array) or a bug in the db40 > > libraries. But the regularity of these errors (I still get them > > from time to time, and have to delete my node.db4o) makes me > > surmise otherwise. > > The error disappeared when I freed some disk space... And I got the > data store shrunk by resizing it; it is at 40% use now. I'll clean > out temp files and node.db4o once the downloads have completed. Interesting. Come to think of it, I also run pretty low on free diskspace ... 99% disk usage (of over 1tb) not being uncommon ... although that should still leave several gigs free. But, it's also interesting that you didn't have to delete your node.db4o file to get rid of the "Internal error". ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
Dennis Nezic wrote: > On Thu, 24 Dec 2009 13:28:36 +0100, SmallSister development wrote: >> 2) Error message on the file sharing pages: >> Internal error >> Internal error >> Return to queue page. >> [The "queue page" gives this error message too] > > I have posted about this bug as far back as June 12th '09, with freenet > version 1216 and probably earlier -- basically ever since the db40 > thing was merged :P. Toad flippantly suggested it was my "flaky > hardware" (raid 5 software array) or a bug in the db40 libraries. But > the regularity of these errors (I still get them from time to time, and > have to delete my node.db4o) makes me surmise otherwise. The error disappeared when I freed some disk space... And I got the data store shrunk by resizing it; it is at 40% use now. I'll clean out temp files and node.db4o once the downloads have completed. ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
Re: [freenet-support] BUG: Freenet does not respect size limit for datastore
On Thu, 24 Dec 2009 13:28:36 +0100, SmallSister development wrote: > > 2) Error message on the file sharing pages: > Internal error > Internal error > Return to queue page. > [The "queue page" gives this error message too] I have posted about this bug as far back as June 12th '09, with freenet version 1216 and probably earlier -- basically ever since the db40 thing was merged :P. Toad flippantly suggested it was my "flaky hardware" (raid 5 software array) or a bug in the db40 libraries. But the regularity of these errors (I still get them from time to time, and have to delete my node.db4o) makes me surmise otherwise. ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
[freenet-support] BUG: Freenet does not respect size limit for datastore
I found that Freenet (Build #1239) happily expands the datastore over the configured size limit. (Disk space is the limit...) This leads to very interesting behaviour when the file system gets full and I would like your help to solve the following two issues: 1) Error message on all pages: The auto-updater may have been compromised! We have turned it off for now, please click for more details. The auto-updater has been disabled. This might be because of a local problem, such as running out of disk space, or the auto-updating system may have been compromised. The reason is: Permanent error fetching revocation (error inserting the revocation key?): FetchException:Temporary files error:null:-1:null:false:null:Failed to close binary blob stream: java.io.IOException: No space left on device. [Bug 2: there is no link to click on at all...] 2) Error message on the file sharing pages: Internal error Internal error Return to queue page. [The "queue page" gives this error message too] ___ Support mailing list Support@freenetproject.org http://news.gmane.org/gmane.network.freenet.support Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support Or mailto:support-requ...@freenetproject.org?subject=unsubscribe