Archiva

unfinished scanning after NPE

Details

  • Type: Bug Bug
  • Status: Closed Closed
  • Priority: Major Major
  • Resolution: Duplicate
  • Affects Version/s: 1.0-beta-3
  • Fix Version/s: None
  • Component/s: repository scanning
  • Labels:
    None
  • Environment:
    macosx 1.4, jdk 1.5, archiva beta3
  • Number of attachments :
    3

Description

I repeatedly tried to index a mirror of central repository. In previous attempts, usually got "Too many files opened" kind of errors. This time it's a NPE, see attached console output

  1. Text File
    error3.txt
    12/Nov/07 1:54 AM
    7 kB
    Milos Kleint
  2. Text File
    error2.txt
    06/Nov/07 1:03 AM
    7 kB
    Milos Kleint
  3. Text File
    error.txt
    05/Nov/07 9:50 AM
    16 kB
    Milos Kleint

Issue Links

duplicates

Bug - A problem which impairs or prevents the functions of the product. MRM-715 error adding artifacts to Lucene under some circumstances

  • Major - Major loss of function.
  • Closed - The issue is considered finished, the resolution is correct. Issues which are not closed can be reopened.

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Milos Kleint added a comment -

yet another attempt to index repository

Show
Milos Kleint added a comment - yet another attempt to index repository
Hide
Permalink
Joakim Erdfelt added a comment -

I've tested this with 1.0-beta-3 and the current archiva/trunk on my full central (ibiblio) mirror.
I cannot reproduce this error.

12384893 [pool-2-thread-1] INFO  org.codehaus.plexus.taskqueue.execution.TaskExecutor:repository-scanning  - Finished repository task:
.\ Scan of ibiblio \.__________________________________________
  Repository Dir    : /home/repo1/ibiblio
  Repository Name   : Ibiblio Mirror
  Repository Layout : default
  Known Consumers   : (5 configured)
                      create-missing-checksums
                      update-db-artifact
                      index-content
                      auto-rename
                      auto-remove
  Invalid Consumers : <none>
  Duration          : 1 Hour 35 Seconds 590 Milliseconds
  When Gathered     : 11/6/07 2:02 PM
  Total File Count  : 281074
  Avg Time Per File : 12 Milliseconds
______________________________________________________________
Show
Joakim Erdfelt added a comment - I've tested this with 1.0-beta-3 and the current archiva/trunk on my full central (ibiblio) mirror. I cannot reproduce this error. 12384893 [pool-2-thread-1] INFO org.codehaus.plexus.taskqueue.execution.TaskExecutor:repository-scanning - Finished repository task: .\ Scan of ibiblio \.__________________________________________ Repository Dir : /home/repo1/ibiblio Repository Name : Ibiblio Mirror Repository Layout : default Known Consumers : (5 configured) create-missing-checksums update-db-artifact index-content auto-rename auto-remove Invalid Consumers : <none> Duration : 1 Hour 35 Seconds 590 Milliseconds When Gathered : 11/6/07 2:02 PM Total File Count : 281074 Avg Time Per File : 12 Milliseconds ______________________________________________________________
Hide
Permalink
Milos Kleint added a comment -

how do I start with a clean table? i've deleted ~/.m2/archiva then the data/archiva folder and .index in the repository directory.
I always need to start archiva once, it complains on the console about missing tables, I shut it down, start again and trigger the scanning..

Show
Milos Kleint added a comment - how do I start with a clean table? i've deleted ~/.m2/archiva then the data/archiva folder and .index in the repository directory. I always need to start archiva once, it complains on the console about missing tables, I shut it down, start again and trigger the scanning..
Hide
Permalink
Brett Porter added a comment -

what you've done should be as clean as you can get, I think.

I don't get the restart problem - the complaints at the start are about SA and JPOX_ tables which can be safely ignored. If you are getting other problems, then there is certainly something going on where you have a database still in use somehow.

looking at the exceptions it seems to indicate a race condition. I can't see what would cause it though - so one thing to try might be to avoid any interaction with the web interface during the initial scan.

We should spend some time investigating this - but I'm inclined to move it post 1.0... wdyt?

Show
Brett Porter added a comment - what you've done should be as clean as you can get, I think. I don't get the restart problem - the complaints at the start are about SA and JPOX_ tables which can be safely ignored. If you are getting other problems, then there is certainly something going on where you have a database still in use somehow. looking at the exceptions it seems to indicate a race condition. I can't see what would cause it though - so one thing to try might be to avoid any interaction with the web interface during the initial scan. We should spend some time investigating this - but I'm inclined to move it post 1.0... wdyt?
Hide
Permalink
Brett Porter added a comment -

I was able to scan 71% of the repository successfully (up to org.apache.something) before I needed to shutdown and leave. It was going very slowly, but successfully. Postponing for further investigation

Show
Brett Porter added a comment - I was able to scan 71% of the repository successfully (up to org.apache.something) before I needed to shutdown and leave. It was going very slowly, but successfully. Postponing for further investigation
Hide
Permalink
Milos Kleint added a comment -

yet again failed scanning after cleaning the data structures

Show
Milos Kleint added a comment - yet again failed scanning after cleaning the data structures
Hide
Permalink
Brett Porter added a comment -

fixed in 1.0.2

Show
Brett Porter added a comment - fixed in 1.0.2

People

  • Assignee:
    Brett Porter
    Reporter:
    Milos Kleint
Vote (2)
Watch (2)

Dates

  • Created:
    Updated:
    Resolved: