**** BEGIN LOGGING AT Fri Mar 13 02:59:57 2009 Mar 13 08:40:43 hmmm ... anyone else seeing a buildfailure while the build run tries to build "setpwc" ? Mar 13 08:41:16 it seems as though the patch stage is trying to execute Before the extract stage (all patch hunks fail with a "No file to patch" message Mar 13 08:41:31 this is trying to build slugosle-5.3-beta Mar 13 10:55:19 Caelian: Yes, the upstream setpwc tarball is corrupted. You'll have to download the setpwc tarball for the 1.1 version from another location to get the correct one, until such time as the upstream folks fix their error. Mar 13 10:57:20 mwester: ah i see :) Mar 13 10:58:21 If you check the md5sum of the 1.1 tarball, and look that up in the checksums.ini file in your openembedded/conf directory, you'll probably note that it's the same md5 as the 0.9 version tarball! Mar 13 11:02:28 btw .. any news on the whole checksumming problem ... Mar 13 11:03:20 no problem there, IMO. It's working great, and has caught several upstream problems. Mar 13 11:03:53 I'm surprised you didn't run into a busybox build failure as well -- they just "refreshed" a patch tarball out from underneath us. Mar 13 11:03:55 i am seeing a consistent problem about the build system not being able to create a checksum object of some sorts Mar 13 11:04:07 You'll have to be more specific. Mar 13 11:04:11 i know Mar 13 11:04:24 letting the build run to catch the exact warning Mar 13 11:05:33 it's something i've been seeing ever since they seemed to have refactored the mechanism to perform the checksum process Mar 13 11:09:00 hmm .. all setpwc-1.1.tgz tarballs i can find contain only an "upgrade.txt" message Mar 13 11:13:26 ok ... going to have to force a package rebuild to trigger the checksum problem i am seeing Mar 13 11:13:42 going to take a little while Mar 13 11:14:08 No big deal, as long as it's building... Mar 13 11:14:59 well .. still can't find a proper setpwc-1.1.tgz ... but that seems to be the only build problem i have encountered so far though Mar 13 11:15:01 You may just have to remove setpwc from the package feeds list. It looks like I'll have to upgrade the feeds, which isn't what I wanted to be doing anytime soon. Mar 13 11:17:16 where exactly is this feeds list maintained ? Mar 13 11:17:32 (this is the first time using the make slugosle-%-beta targets( Mar 13 11:19:37 in releases/openembedded/packages/meta/slugos_packages.bb (or similar filename) Mar 13 11:19:59 aah ok .. makes sense .. i'll check there :) Mar 13 11:20:08 BTW, do you really need the entire package feeds locally? or just the image? Mar 13 11:20:31 mwester: generally just the image .. but it was nice to be able to do a full package feed locally Mar 13 11:20:47 Ok. Mar 13 11:21:12 it's not unlikely i'll be needing a local feed to persist for a long time Mar 13 11:21:35 It should work either way; we're just too close to the "bleeding edge" for a lot of stuff for this release. It's unfortunate that we're only a few days from the release, and already two packages won't build due to upstream changes. Mar 13 11:22:50 * Caelian kicks his old build system ... hurry up already :) Mar 13 11:23:23 * mwester can build the full release in a little over 2 hours. Mar 13 11:23:43 on a more powerful system i am sure i could as well :) Mar 13 11:23:58 but it takes the better part of a day on the official build server, which is really quite a pain for releasing. Mar 13 11:24:01 but i'm using an old otherwise considered "junk" system for the process Mar 13 11:25:40 damned ... looks like i forgot to whipe the staging package Mar 13 11:29:26 and all i want is to trigger the checksum problem i am seeing that's seemingly causing checksums to simply be ignored Mar 13 11:29:38 (which is likely why i didn't notice the busybox breakage) Mar 13 11:30:50 Ah, ok - I misunderstood your issue. Yes, there's a problem that results in checksums being ignored for some recipes. Mar 13 11:31:11 for me it's ignoring checksums across the board it seems Mar 13 11:31:16 That's been fixed in the mainline already. Mar 13 11:31:40 Ok, it's order-sensitive IIRC, so I think it depends on what gets built when. Mar 13 11:35:41 the error is "Creating the CheckSum parser failed" Mar 13 11:37:15 it doesn't cause the build to abort ... but obviously checksums mismatches go undetected this way Mar 13 11:37:25 and i don't understand why this happens Mar 13 11:38:00 Yep, that's the one. You can check the git history on the OE dev branch to see what the bug was and how they fixed it. Mar 13 11:38:32 aah great Mar 13 11:40:57 not that versed in git and oe yet ... but good to know the problem has been resolved in oe Mar 13 11:45:25 aaah ... looks like a simple superfluous control characters at the beginning of openembedded/conf/checksums.ini Mar 13 21:58:06 03bzhou * r9729 10optware/trunk/make/cherokee.mk: cherokee: 0.99.3 -> 0.99.4 Mar 13 22:48:22 hello there... can anyone help me compile appweb ??? Mar 13 22:48:46 i get an error in ejs, and well i dont know ejs verry well Mar 13 22:48:59 i am using debian lenny ..btw... Mar 14 02:49:54 03bzhou * r9730 10optware/trunk/make/libidn.mk: libidn: 1.10 -> 1.13 Mar 14 02:49:55 03bzhou * r9731 10optware/trunk/make/gloox.mk: gloox: 0.9.9.5 -> 0.9.9.7 **** ENDING LOGGING AT Sat Mar 14 02:59:57 2009