[FOSSology] filter_license job failing

2009-09-03 Thread Dave McLoughlin
I just recently did a clean install of v1.1 on Ubuntu.  I did some
license analysis on a couple files I uploaded from url and the
analysis completed fine.

 

Now I am trying to do some analysis on some files I uploaded from
server and the analysis keeps failing at the filter_license stage.

 

I'm attaching my log file from the point I restarted Fossology upto the
latest failure.

 

Please let me know if you can tell me why the job is failing.

 


Dave McLoughlin |  OpenLogic



 



fossology.log
Description: fossology.log
___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] filter_license job failing

2009-09-03 Thread Dave McLoughlin
Great, thanks for the info. Yes, I did see the stack smashing detected
in my stdout.  Unfortunately, I didn't realize it was related to my
issue.  I should have read it more carefully.

$ *** stack smashing detected ***:
/usr/local/lib/fossology/agents/Filter_License terminated 
=== Backtrace: =
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb7effda8]
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb7effd60]
/usr/local/lib/fossology/agents/Filter_License[0x804d38d]
/usr/local/lib/fossology/agents/Filter_License[0x804e1e3]
/usr/local/lib/fossology/agents/Filter_License[0x804b68c]
/usr/local/lib/fossology/agents/Filter_License[0x804d84a]
/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7e18775]
/usr/local/lib/fossology/agents/Filter_License[0x80498f1]
=== Memory map: 
08048000-08055000 r-xp  08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08055000-08056000 r--p c000 08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08056000-08057000 rw-p d000 08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08057000-0805a000 rw-p 08057000 00:00 0 
08d67000-08e5d000 rw-p 08d67000 00:00 0  [heap]

Etc...


I'll try your workaround.  Thanks, you saved me a bunch of time.


Dave McLoughlin |  OpenLogic

720 240 4530  |  phone
303 818 1686  |  cell 


-Original Message-
From: Dragoslav Mitrinovic [mailto:dr...@motorola.com] 
Sent: Thursday, September 03, 2009 1:02 PM
To: Dave McLoughlin
Cc: fossology@fossology.org
Subject: Re: [FOSSology] filter_license job failing

Dave,

I've reported similar issue in Ubuntu about a month ago, see thread
starting at this message: 
http://www.mail-archive.com/fossology@fossology.org/msg00545.html

You could confirm you are seeing the same issue by capturing the agent
stdout/stderr and seeing if Filter_License is crashing due to stack
smashing. (I captured agent stderr/stdout by restarting fossology
service from shell command line.)

The last message in the thread contains my diagnosis of the likely
cause. What I did NOT report to the list was my workaround: I recompiled
Filter_License agent with stack smashing protection disabled, by adding 
-fno-stack-protector into CFLAGS_LOCAL in
agents/foss_license_agent/Filter_License/Makefile, and Filter_License
has been running fine ever since. 

Of course, your issue may be something totally different, in which case
someone more qualified than myself will have to help.

Best regards,
-Dragoslav



On Thu, 2009-09-03 at 11:56 -0600, Dave McLoughlin wrote:
 I just recently did a clean install of v1.1 on Ubuntu.  I did some
 license analysis on a couple files I uploaded from url and the
 analysis completed fine.
 
  
 
 Now I am trying to do some analysis on some files I uploaded from
 server and the analysis keeps failing at the filter_license stage.
 
  
 
 I'm attaching my log file from the point I restarted Fossology upto
 the latest failure.
 
  
 
 Please let me know if you can tell me why the job is failing.
 
  
 
 
 Dave McLoughlin |  OpenLogic
 
 
 
  
 
 
 ___
 fossology mailing list
 fossology@fossology.org
 http://fossology.org/mailman/listinfo/fossology

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] filter_license job failing

2009-09-03 Thread Dave McLoughlin
Dan,

Is there any other information I can supply to help?  Let me know. 

BTW, I'm attaching the complete dump from my stdout in case you find it
useful.


Dave McLoughlin |  OpenLogic



-Original Message-
From: Dan Stangel [mailto:dan...@fossology.org] 
Sent: Thursday, September 03, 2009 2:36 PM
To: Dave McLoughlin; Dragoslav Mitrinovic
Cc: fossology@fossology.org
Subject: RE: [FOSSology] filter_license job failing

A workaround is nice, but we really ought to fix the stack overflow
issue
itself in filter_license, yes (assuming that is what's happening)?

Dan

-Original Message-
From: fossology-boun...@fossology.org
[mailto:fossology-boun...@fossology.org] On Behalf Of Dave McLoughlin
Sent: Thursday, September 03, 2009 1:14 PM
To: Dragoslav Mitrinovic
Cc: fossology@fossology.org
Subject: Re: [FOSSology] filter_license job failing

Great, thanks for the info. Yes, I did see the stack smashing detected
in my stdout.  Unfortunately, I didn't realize it was related to my
issue.  I should have read it more carefully.

$ *** stack smashing detected ***:
/usr/local/lib/fossology/agents/Filter_License terminated 
=== Backtrace: =
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb7effda8]
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb7effd60]
/usr/local/lib/fossology/agents/Filter_License[0x804d38d]
/usr/local/lib/fossology/agents/Filter_License[0x804e1e3]
/usr/local/lib/fossology/agents/Filter_License[0x804b68c]
/usr/local/lib/fossology/agents/Filter_License[0x804d84a]
/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7e18775]
/usr/local/lib/fossology/agents/Filter_License[0x80498f1]
=== Memory map: 
08048000-08055000 r-xp  08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08055000-08056000 r--p c000 08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08056000-08057000 rw-p d000 08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08057000-0805a000 rw-p 08057000 00:00 0 
08d67000-08e5d000 rw-p 08d67000 00:00 0  [heap]

Etc...


I'll try your workaround.  Thanks, you saved me a bunch of time.


Dave McLoughlin |  OpenLogic

720 240 4530  |  phone
303 818 1686  |  cell 


-Original Message-
From: Dragoslav Mitrinovic [mailto:dr...@motorola.com] 
Sent: Thursday, September 03, 2009 1:02 PM
To: Dave McLoughlin
Cc: fossology@fossology.org
Subject: Re: [FOSSology] filter_license job failing

Dave,

I've reported similar issue in Ubuntu about a month ago, see thread
starting at this message: 
http://www.mail-archive.com/fossology@fossology.org/msg00545.html

You could confirm you are seeing the same issue by capturing the agent
stdout/stderr and seeing if Filter_License is crashing due to stack
smashing. (I captured agent stderr/stdout by restarting fossology
service from shell command line.)

The last message in the thread contains my diagnosis of the likely
cause. What I did NOT report to the list was my workaround: I recompiled
Filter_License agent with stack smashing protection disabled, by adding 
-fno-stack-protector into CFLAGS_LOCAL in
agents/foss_license_agent/Filter_License/Makefile, and Filter_License
has been running fine ever since. 

Of course, your issue may be something totally different, in which case
someone more qualified than myself will have to help.

Best regards,
-Dragoslav



On Thu, 2009-09-03 at 11:56 -0600, Dave McLoughlin wrote:
 I just recently did a clean install of v1.1 on Ubuntu.  I did some
 license analysis on a couple files I uploaded from url and the
 analysis completed fine.
 
  
 
 Now I am trying to do some analysis on some files I uploaded from
 server and the analysis keeps failing at the filter_license stage.
 
  
 
 I'm attaching my log file from the point I restarted Fossology upto
 the latest failure.
 
  
 
 Please let me know if you can tell me why the job is failing.
 
  
 
 
 Dave McLoughlin |  OpenLogic
 
 
 
  
 
 
 ___
 fossology mailing list
 fossology@fossology.org
 http://fossology.org/mailman/listinfo/fossology

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology




$ *** stack smashing detected ***: 
/usr/local/lib/fossology/agents/Filter_License terminated 
=== Backtrace: = 
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb7effda8]
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb7effd60]
/usr/local/lib/fossology/agents/Filter_License[0x804d38d]
/usr/local/lib/fossology/agents/Filter_License[0x804e1e3]
/usr/local/lib/fossology/agents/Filter_License[0x804b68c]
/usr/local/lib/fossology/agents/Filter_License[0x804d84a]
/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7e18775]
/usr/local/lib/fossology/agents/Filter_License[0x80498f1]
=== Memory map: 
08048000-08055000 r-xp  08:01 2778370
/usr/local/lib/fossology/agents/Filter_License
08055000-08056000 r--p c000 08

[FOSSology] '/cp2foss-4ac248402ea7c.tar' does not exist.

2009-09-29 Thread Dave McLoughlin
I was just trying to upload a directory for analysis with Fossology 1.0
on Ubuntu.  The upload failed with an error that the cp2foss does not
exist.

 

Here is an excerpt from the log:

 

 


Dave McLoughlin |  OpenLogic
dave.mclough...@openlogic.com

720 240 4530  |  phone
303 818 1686  |  cell 
720 240 4556  |  fax
@OpenLogic|  twitter
1 888 OpenLogic  |  toll free
www.openlogic.com

OpenLogic, Inc.
Headquarters, Broomfield, Colorado 80021

  http://www.openlogic.com 

 

image001.gif___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] Moving the repository

2010-11-03 Thread Dave McLoughlin
I have been off the mailing list for some time, so I apologize in advance if
this has already been answered.  What is the best way to move a repository
to a new system?  I installed Fossology on a different machine, did a
database dump and imported into the new install fine.  But I haven¹t had any
luck getting Fossology to see my repository that I moved to the new system.

Here are the steps I took:
1. installed Fossology on a new computer
2. dumped the original install database
3. imported the database dump into postgresql on the new computer (at this
point I was able to browse all my uploads)
4. tar¹ed up the repository directory
5. moved it to the new computer (I put it in a different location so I
updated my RepPath.conf file)
6. when I first moved the repository the user and group were set to root so
I updated it to be fossy:fossy

At this point when I go to view a file in fossology I get a message that the
file is not in the repository.

I¹m using localhost on both systems, so I didn¹t update my Hosts.conf.

Thanks,


-- 
Dave McLoughlin |  OpenLogic



___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Install of 1.2.1

2010-11-15 Thread Dave McLoughlin
I¹m using the source.

dave


On 11/15/10 11:35 AM, Mark Donohoe mark.dono...@hp.com wrote:

 Dave,
 
 Installing from source on 9.10 should work.  Did you use the packages or
 source?
 It appears you tried to upgrade from 1.1.0, which makes sense. I'll start
 looking into this.
 
 
 Dave McLoughlin wrote:
Install of 1.2.1
  
 
 Hi folks,
  
 I'm running into a few issue with installing (upgrading, I had 1.10
 installed) Fossology 1.2.1 on Ubuntu and was hoping someone might be able to
 assist me.
  
 issues:
 1) fo-installdeps - I get an error that I have an Unknown or Unsupport
 version of Ubuntu.  I have Ubuntu 9.10 installed.  Here are the last few
 lines of the console:
  
 The following packages were automatically installed and are no longer
 required:
   libstdc++6-4.3-dev linux-headers-2.6.28-15 g++-4.3 phonon
   libmysqlclient15off phonon-backend-gstreamer libmagick++1 libphonon4
   libkadm55 librsvg2-dev postgresql-8.3 linux-headers-2.6.28-15-generic
   qt4-qtconfig
 Use 'apt-get autoremove' to remove them.
 0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded.
 ERROR: Unknown or Unsupported Ubuntu release
  
 2) After the install I'm see the following in my browser when I navigate to
 http://localhost/repo/:
  
 Warning: file_get_contents(/usr/local/etc/fossology/Db.conf)
 [function.file-get-contents]: failed to open stream: Permission denied in
 /usr/local/share/fossology/www/plugins/core-db.php on line 91
  
 Warning: pg_pconnect() [function.pg-pconnect]: Unable to connect to
 PostgreSQL server: FATAL: database www-data does not exist in
 /usr/local/share/fossology/www/plugins/core-db.php on line 91
  
 Warning: pg_last_error(): supplied argument is not a valid PostgreSQL link
 resource in /usr/local/share/fossology/www/plugins/core-db.php on line 136
  
 The user/group is set to root on those files.  Do they need to fossy/fossy?
  
 Any assistance would be appreciated.
  
 thanks
  
   
  
 
 
 
 ___
 fossology mailing list
 fossology@fossology.org
 http://fossology.org/mailman/listinfo/fossology
   
 

-- 
Dave McLoughlin |  OpenLogic
dave.mclough...@openlogic.com

720 240 4530  | phone
303 818 1686  | cell
720 240 4556  | fax
@OpenLogic| twitter
1 888 OpenLogic  | toll free
www.openlogic.com
olex.openlogic.com

OpenLogic, Inc.
Headquarters, Broomfield, Colorado 80021

 

image.gif___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Announcing FOSSology 1.3.0 Release (Laser, Mary)

2011-01-20 Thread Dave McLoughlin
I found it.  Here¹s what I was referring to:

From Mary Laser to the fossology mailing list on 11/12/10:

³Furthermore, there is NO upgrade path from bSAM results to nomos results.
Therefore, you MUST rerun license analysis on your uploads using the nomos
analyzer introduced in 1.2.0.²

Thanks,
-- 
Dave McLoughlin |  OpenLogic

 

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Announcing FOSSology 1.3.0 Release (Laser, Mary)

2011-01-20 Thread Dave McLoughlin
I¹m not sure I understand what ³upgrade path from bsam results to nomos
results² means.

If I have some scans from 1.1 and I upgrade to 1.3, does it mean I won¹t be
able to view the old results unless I rescan them in 1.2 using nomos.

re: suggestions? I don¹t want to have to rescan everything I currently have
in Fossology that was done using bSam.  I would like to be able to at a
minimum still be able to see the old results.  When you remove the bSam
code, does that mean you also remove the ability to see any old data that
was originally scanned with bSam?

Thanks,

Dave


On 1/20/11 1:04 PM, Bob Gobeille bob.gobei...@hp.com wrote:

 This is correct.  Since we get better and faster results with nomos, and bSAM
 is difficult to maintain, we no longer want to support it.  We don't want to
 migrate your data because it is potentially very time consuming to rescan all
 your uploads.   This would be especially wasteful if you are no longer
 interested in all of your previously scanned data.
 
 We have been putting out messages about this for quite a while but have
 receive no feedback until your email arrived.   All the bsam code is still in
 1.3 but it is no longer supported.  In the next release we would like to
 remove that code, which will orphan your bsam data.
 
 Suggestions?
 
 Bob Gobeille
 
 
 On Jan 20, 2011, at 12:20 PM, Dave McLoughlin wrote:
 
 I found it.  Here¹s what I was referring to:
 
 From Mary Laser to the fossology mailing list on 11/12/10:
 
 ³Furthermore, there is NO upgrade path from bSAM results to nomos results.
 Therefore, you MUST rerun license analysis on your uploads using the nomos
 analyzer introduced in 1.2.0.²
 
 

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] Fossology on a Mac

2011-01-27 Thread Dave McLoughlin
Has anyone tried to install fossology on a Mac OS X system?  Just curious.

-- 
Dave McLoughlin |  OpenLogic

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Upgrade path for 1.3.0 (Mark Donohoe)

2011-02-03 Thread Dave McLoughlin
Mark,

Thanks for the update on 1.3.1.  I don't see needing bsam or the other
features you are removing going forward.  Thanks for keeping access to the
previously created bsam data, though.

This will definitely meet our needs.

Dave


On 2/3/11 12:00 PM, fossology-requ...@fossology.org
fossology-requ...@fossology.org wrote:

 Send fossology mailing list submissions to
 fossology@fossology.org
 
 To subscribe or unsubscribe via the World Wide Web, visit
 http://fossology.org/mailman/listinfo/fossology
 or, via email, send a message with subject or body 'help' to
 fossology-requ...@fossology.org
 
 You can reach the person managing the list at
 fossology-ow...@fossology.org
 
 When replying, please edit your Subject line so it is more specific
 than Re: Contents of fossology digest...
 
 
 Today's Topics:
 
1. Upgrade path for 1.3.0 (Dave McLoughlin)
2. Re: Upgrade path for 1.3.0 (Mark Donohoe)
 
 
 --
 
 Message: 1
 Date: Wed, 02 Feb 2011 12:20:49 -0700
 From: Dave McLoughlin dave.mclough...@openlogic.com
 Subject: [FOSSology] Upgrade path for 1.3.0
 To: fossology@fossology.org fossology@fossology.org
 Message-ID: c96efca1.6778%dave.mclough...@openlogic.com
 Content-Type: text/plain; charset=iso-8859-1
 
 Folks,
 
 I?m embarrassed to say I?m still on 1.1.0, but am ready to go to 1.3.0  I
 actually tested an install of 1.2.1 on a vm and didn?t have any issues.  I
 just put it on top of the 1.1.0 I had installed in the vm and it worked
 fine, i.e. all the data from previous scans were still there and accessible.
 
 When I got to 1.3.0 do I need to do an intermediate build/install of 1.2.0
 first?  Or, can I go right to 1.3.0?  I would like to keep all the old scans
 I have in 1.1.0.
 
 Thanks,
 
 Dave
 -- next part --
 An HTML attachment was scrubbed...
 URL: 
 http://fossology.org/pipermail/fossology/attachments/20110202/214b6501/attach
 ment.html
 
 --
 
 Message: 2
 Date: Wed, 02 Feb 2011 13:33:55 -0800
 From: Mark Donohoe mark.dono...@hp.com
 Subject: Re: [FOSSology] Upgrade path for 1.3.0
 To: fossology@fossology.org
 Message-ID: 4d49cdc3.2010...@hp.com
 Content-Type: text/plain; charset=iso-8859-1; Format=flowed
 
 On 02/02/2011 11:20 AM, Dave McLoughlin wrote:
 Folks,
 
 I'm embarrassed to say I'm still on 1.1.0, but am ready to go to 1.3.0
  I actually tested an install of 1.2.1 on a vm and didn't have any
 issues.  I just put it on top of the 1.1.0 I had installed in the vm
 and it worked fine, i.e. all the data from previous scans were still
 there and accessible.
 
 When I got to 1.3.0 do I need to do an intermediate build/install of
 1.2.0 first?  Or, can I go right to 1.3.0?  I would like to keep all
 the old scans I have in 1.1.0.
 
 Thanks,
 
 Dave
 Dave,
 We don't really support 'jumping' releases.  By that I mean that we have
 never tested it.  The safest way to upgrade to 1.3.0 would be to install
 (or build/install if installing from source) 1.2.1, not use it, and then
 upgrade to 1.3.0.  In this way you are sure to get all the db changes
 between the releases.  None of our releases ever remove user data from
 the DB.  That would be a serious defect.
 
 As an FYI, for 1.3.1, bsam, license groups, license terms  and the Spec
 agent are being removed from the product.  We are keeping the ability to
 display bsam results, but no more basm analysis (agents) will be
 available after 1.3.0.  We believe this will meet your needs for the
 display of older basm results.  Does this meet your needs?
 
 ---
 Mark Donohoe
 OSPO, HP
 FOSSology.org
 
 -- next part --
 An HTML attachment was scrubbed...
 URL: 
 http://fossology.org/pipermail/fossology/attachments/20110202/90bc1638/attach
 ment-0001.htm
 
 --
 
 ___
 fossology mailing list
 fossology@fossology.org
 http://fossology.org/mailman/listinfo/fossology
 
 
 End of fossology Digest, Vol 40, Issue 3
 

-- 
Dave McLoughlin |  OpenLogic
dave.mclough...@openlogic.com

720 240 4530  | phone
303 818 1686  | cell
720 240 4556  | fax
@OpenLogic| twitter
1 888 OpenLogic  | toll free
www.openlogic.com
olex.openlogic.com

OpenLogic, Inc.
Headquarters, Broomfield, Colorado 80021

 

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] 1.3.0 Install issues

2011-02-08 Thread Dave McLoughlin

I completed the upgrade from 1.1 to 1.2.1 this morning and the install worked 
fine.  I was able to view all existing scans in my database.  And I ran a new 
sample scan and it worked fine.

I just attempted to upgrade to 1.3.0 and saw a few errors.  I was wondering 
if you can help me debug.

during the make install, I'm seeing:

make[2]: Entering directory `/home/dave/fossology-1.3.0/agents/copyright'
/usr/bin/install -D copyright /usr/local/lib/fossology/agents/
/usr/bin/install: cannot overwrite directory 
`/usr/local/lib/fossology/agents/copyright' with non-directory
make[2]: *** [install] Error 1
make[2]: Leaving directory `/home/dave/fossology-1.3.0/agents/copyright'
make[1]: *** [install-copyright] Error 2
make[1]: Leaving directory `/home/dave/fossology-1.3.0/agents'
make: *** [install-agents] Error 2

Then when I run the fossology-scheduler test it fails:

$ sudo /usr/local/lib/fossology/fossology-scheduler -t
Exec failure reason: Permission denied

*** StopScheduler DELETE from scheduler_status. Status ,  ***

Logs say: 

2011-02-08 12:06:05 scheduler[18844] : STATUS: 1 agents failed to initialize.
2011-02-08 12:06:05 scheduler[18844] : *** 1 agent failures.  Scheduler 
exiting. 


Any help would be appreciated.

Dave
___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] Session time out

2011-02-17 Thread Dave McLoughlin
Fossology logs me out after a few minutes.  Is there a way to configure it
so it doesn¹t time out and just leaves me logged in?

Dave
___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] 1.3.0 feedback

2011-02-18 Thread Dave McLoughlin
I was finally able to get my Nomos License list to display on the screen.  I
had to edit my php.ini max_execution_time.  Default is 90 seconds, I upped
it to 900 and I was able to get it to work.  The nomos list had over 106,000
files.  The problem was that the error didn't always display when it failed.
When it did show up again it said something like Fatal error: max execution
time exceeded.  I'm surprised this message went to the browser and not to
the log files.  Must be another setting someplace else.

The approach that fossology took in regards to displaying the bsam license
tree results, was to feed the data as it became available, so you could see
partial results over time.  It looks like with the nomos license list, you
run a script then pass the results to the browser after all data is
collected.  The prior approach seems better to me because you get a sense of
progress or status.  But with the nomos list approach, you don't really know
if it's working or not.  Does that make sense?

Also, one thing I really liked about the bsam license tree, was that it
included links in the data back to the file.  But the nomos list is just a
static list of files that you have to manually navigate to in order to
review the results.

-- 
Dave McLoughlin |  OpenLogic
dave.mclough...@openlogic.com

720 240 4530  | phone
303 818 1686  | cell
720 240 4556  | fax
@OpenLogic| twitter
1 888 OpenLogic  | toll free
www.openlogic.com
olex.openlogic.com

OpenLogic, Inc.
Headquarters, Broomfield, Colorado 80021

 

image.gif___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Database connection error

2011-07-08 Thread Dave McLoughlin
Mary,

I should have included the postgresql logs yesterday.  Does this help pinpoint 
the issue? This is the last couple entries from the postgresql log:

2011-07-07 15:39:15 MDT [local] 4e162783.ccbLOG:  provided username (fossy) and 
authenticated username (user) don't match
2011-07-07 15:39:15 MDT [local] 4e162783.ccbFATAL:  Ident authentication failed 
for user fossy

thanks,

Dave


-Original Message-
From: Laser, Mary [mailto:mary.la...@hp.com]
Sent: Thu 7/7/2011 3:59 PM
To: Dave McLoughlin; fossology@fossology.org
Subject: RE: Database connection error
 
This usually is a symptom of a mis-configured postgres file; typically 
/etc/postgresql/8.3/main/pg_hba.conf

Look in the postgres log file (/var/log/postgresql/postgresql-8.3-main.log) for 
additional clues to your error.

 

Mary

 

 

From: fossology-boun...@fossology.org [mailto:fossology-boun...@fossology.org] 
On Behalf Of Dave McLoughlin
Sent: Thursday, July 07, 2011 3:46 PM
To: fossology@fossology.org
Subject: [FOSSology] Database connection error

 

I'm running into a problem with a database connection error after install of 
1.4.1 on Ubuntu 1.4.1.

When I re-ran fo-postinstall to make sure everything was okay I got:

*** Running postinstall for everything ***
*** Setting up the FOSSology database ***
NOTE: fossology database already exists, not creating
*** Checking for plpgsql support ***
NOTE: plpgsql already exists in fossology database, good
*** Creating user and group ***
NOTE: group 'fossy' already exists, good.
NOTE: user 'fossy' already exists, good.
*** Making sure needed dirs exist with right ownership/permissions ***
NOTE: Repository already exists at /srv/fossology/repository
NOTE: /srv/fossology/repository/localhost directory already exists
*** Checking database connectivity ***
NOTE: Connect succeeded, database is working.


But my fossology.log shows
2011-07-07 13:30:49 scheduler[1030] : FATAL: fo_watchdog unable to connect to 
database.  Terminating.
2011-07-07 13:34:30 scheduler[2408] : Log opened
2011-07-07 13:34:30 scheduler[2408] : Scheduler kill requested.  Killing 
scheduler.

Also trying psql I get:
$ psql -d fossology -U fossy
psql: FATAL:  Ident authentication failed for user fossy

Any ideas?

Dave

Any ideas?




___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


Re: [FOSSology] Database connection error

2011-07-08 Thread Dave McLoughlin
Mark,

I double checked and the postgresql.conf file has the changes listed in the
INSTALL doc.  I've also made the other listed config changes.

I had tried to do a deb install first, but couldn't get it to work.  I have
Ubuntu 11.04 and it looks like the deb packages don't support that version
of the OS.  So I cleaned everything off and build it from source.  The
fo-installdeps did not work and gave an error message that said Unsupported
version of Ubuntu.  So I manually installed the dependencies with Synaptic
package manager.

The pg_hba.conf is identical to the one I have on an older version of Ubuntu
where I have fossology 1.3.0 installed and running fine.

So I'm not sure where the problem is at this point.

Thanks,

Dave

 Message: 3
 Date: Thu, 07 Jul 2011 15:08:58 -0700
 From: Mark Donohoe mark.dono...@hp.com
 Subject: Re: [FOSSology] Database connection error
 To: fossology@fossology.org
 Message-ID: 4e162e7a.7060...@hp.com
 Content-Type: text/plain; charset=iso-8859-1; Format=flowed
 
 Dave,
 
 When doing a package install, the system prep is not much different that
 with a source install.  That is fossology doesn't want to just go change
 your php.ini files, or postgres files as part of the install.  So either
 before installing the packages or right after (stop the scheduler if you
 have installed packages).
 
 1. Configure kernal shmmax
 2. configure php ini files in cli and apache areas of php
 3. configure postgres
 4. configure apache
 5. restart both apache and postgres
 
 This is explained in the INSTALL document in the sources. Hope that
 helps.  Until all of the above steps have been done, you may continue to
 have problems.  Hope that helps.
 
 
 On 07/07/2011 02:59 PM, Laser, Mary wrote:

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology


[FOSSology] Ubuntu 11.04

2011-08-10 Thread Dave McLoughlin
Has anyone tried to install Fossology 1.4.0 on Ubuntu 11.04?

I ran into some problems a few weeks ago with some database issues and
haven¹t had time to get back to it until this week.  I just tried a fresh
install from source and got the following at the end of fo-postinstall
script:

*** Checking database connectivity ***
NOTE: Connect succeeded, database is working.
*** Initializing database tables ***
PHP Fatal error:  Call to undefined function pg_connect() in
/usr/local/share/fossology/www/common/common-db.php on line 51
ERROR: Unable to initialize.

Mary L had recommended I check out my postgresql config, probably still the
same issue.  I¹m just wondering if anyone has seen this and has been able to
solve it.

Thanks,

Dave

___
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology