Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!

Bug 298306

Summary: chemsum of www-client/seamonkey-2.0.1 failed
Product: Gentoo Linux Reporter: callmewhatyoulike <black-hole>
Component: Current packagesAssignee: Lars Wendler (Polynomial-C) (RETIRED) <polynomial-c>
Status: RESOLVED DUPLICATE    
Severity: normal CC: mozilla
Priority: High    
Version: unspecified   
Hardware: AMD64   
OS: Linux   
Whiteboard:
Package list:
Runtime testing required: ---

Description callmewhatyoulike 2009-12-25 12:05:10 UTC
as in 298305
checksum for seamonkey-2.0.1 (~amd64) failed during fetch. Different servers, resynced from a different mirror, tried to --fetch again, error stays. Invalid checksum. (file is seamonkey-2.0.1.source.tar.bz2)

expected: should just not complain, maybe checksum is broken

possible solution: somebody having a look at the checksums
Comment 1 Lars Wendler (Polynomial-C) (RETIRED) gentoo-dev 2009-12-25 13:35:27 UTC
Did you remove the source file from your ${DISTDIR} before you re-fetched the file?
What checksums are written in the Manifest file in your portage-tree? I have the following ones:

# grep "2\.0\.1\.source" $(portageq portdir)/www-client/seamonkey/Manifest
DIST seamonkey-2.0.1.source.tar.bz2 60965775 RMD160 e9c1953b4e27099fad2fce4d076e694267b6b5e8 SHA1 4469d68d8a2b5db125d600818b504d5b6c133302 SHA256 fcf59a5c8eebb81c4c4db416d4995d3d13505bf47dc4a5dd7a931fda8e85715c

I just removed the source file here, resynced and re-fetched the file without any checksum-errors...
Comment 2 Jory A. Pratt gentoo-dev 2009-12-25 18:03:54 UTC
Resolving distfiles.gentoo.org... 149.20.20.135, 140.211.166.134, 130.239.17.6, ...
Connecting to distfiles.gentoo.org|149.20.20.135|:80... connected.
HTTP request sent, awaiting response... 302 Found
Location: http://mirrors.kernel.org/gentoo/distfiles/seamonkey-2.0.1.source.tar.bz2 [following]
--2009-12-25 12:00:50--  http://mirrors.kernel.org/gentoo/distfiles/seamonkey-2.0.1.source.tar.bz2
Resolving mirrors.kernel.org... 149.20.20.135, 204.152.191.39
Reusing existing connection to distfiles.gentoo.org:80.
HTTP request sent, awaiting response... 200 OK
Length: 60965775 (58M) [application/x-bzip2]
Saving to: `/usr/portage/distfiles/seamonkey-2.0.1.source.tar.bz2'

100%[======================================>] 60,965,775   699K/s   in 2m 18s  

2009-12-25 12:03:09 (430 KB/s) - `/usr/portage/distfiles/seamonkey-2.0.1.source.tar.bz2' saved [60965775/60965775]

 * seamonkey-2.0.1.source.tar.bz2 RMD160 SHA1 SHA256 size ;-) ...        [ ok ]

Bug needs to be handed off to mirror admins. report will need to provide which mirrors he was been trying to get the file off of.
Comment 3 callmewhatyoulike 2009-12-26 11:43:10 UTC
(In reply to comment #1)
> Did you remove the source file from your ${DISTDIR} before you re-fetched the
> file?
a) the files were automatically renamed by the fetch to "seamonkey-something...failed"
b) yes, later I cleared the whole /usr/portage/distfiles/* but it still failed.

Manifest (no resynced yet) says:
cat /usr/portage/www-client/seamonkey/Manifest
[...]many patches, langfiles etc.[...]
DIST seamonkey-2.0.1.source.tar.bz2 60965775 
RMD160 e9c1953b4e27099fad2fce4d076e694267b6b5e8
SHA1 4469d68d8a2b5db125d600818b504d5b6c133302
SHA256 fcf59a5c8eebb81c4c4db416d4995d3d13505bf47dc4a5dd7a931fda8e85715c

it failed with the RMD160 with an incorrect checksum for the downloaded file. Same was for mplayer (bug #298305). Strange is that we seem to have the same checksums. Other files downloaded worked well. (my whole -fu world just failed on seamonkey and mplayer with many different servers (even Mozilla's own server)). So I have no clue what's wrong. 
I'll resync now and see what happens.
Comment 4 callmewhatyoulike 2009-12-26 12:00:29 UTC
Resynced
and: emerge -f seamonkey (~amd64)

HTTP request sent, awaiting response... 200 OK
Length: 60965775 (58M) [application/x-bzip2]
Saving to: `/usr/portage/distfiles/seamonkey-2.0.1.source.tar.bz2'
100%[=====================================================================================================================================================================>] 60,965,775  1.71M/s   in 35s

2009-12-26 13:54:07 (1.68 MB/s) - `/usr/portage/distfiles/seamonkey-2.0.1.source.tar.bz2' saved [60965775/60965775]

('Failed on RMD160 verification', '4e9d9545f3355c4510db00ec1acc450941af4af1', 'e9c1953b4e27099fad2fce4d076e694267b6b5e8')
!!! Fetched file: seamonkey-2.0.1.source.tar.bz2 VERIFY FAILED!
!!! Reason: Failed on RMD160 verification
!!! Got:      4e9d9545f3355c4510db00ec1acc450941af4af1
!!! Expected: e9c1953b4e27099fad2fce4d076e694267b6b5e8
Refetching... File renamed to '/usr/portage/distfiles/seamonkey-2.0.1.source.tar.bz2._checksum_failure_.Nv9kX2'

On an experiment I just noticed while some files work nicely another large file (vanilla-sources) failed, too. That makes me wonder if my NIC has gone stupid or if there might be a driver issue. (one of these Atheros LAN chips on an asus mainboard (marked experimental in kernel since 2.6.29))
I'm really sorry if it was my HW's fault that corrupted the downloads and made me bother the devs. *ashamed*
I'll go and check that with a different NIC.
Comment 5 callmewhatyoulike 2009-12-26 17:51:20 UTC
OMG. This is so embarassing. My NIChip failed, or it got messed with the driver in 2.6.31-g-r1. I tried from a different hardware: different box, also Gentoo and it worked right away. Tried the problematic box but with an additional NIC and suddenly the fetch/download of files above 50MB wouldn't corrupt. 
Sorry for bothering you people and shame on me for not researching my box closely.
But shame even more on ASUS and Atheros with their high priced but buggy chips and bios updates. (This is not the 1st time that the ath chip did strange things on that board.) 
And praise the good old ever working RTL8139x.
Sorry for the inconvenience and trouble I caused. I'll mark it as solved/invalid.

*** This bug has been marked as a duplicate of bug 298305 ***