Bugs item #1792353, was opened at 2007-09-11 15:30
Message generated for change (Settings changed) made by mlkersten
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1792353&group_id=56967

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Core
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Robin Aly (raly)
Assigned to: Nobody/Anonymous (nobody)
>Summary: XQ:Crash while updating XML

Initial Comment:
The problem occured of "time wise"-joining mpeg7 documents. There are shot 
segments and speaker segments both having a start and a duration. Now I want to 
update the shot segments <mpeg7:VideoSegment> with all text data from the 
overlapping speaker segments. I shred with pf:add-doc(PATH, BASENAME, 
"trecvid2007,10). This is done for ca 130 docs. Then I execute integrate.xq 
(see attachment) on each of them by substituting THEDUMMY with the actual 
filename. For the first 10 docs this works then on 120.mp7.xml it stops: " 
cannot mmap(): name=20/2042, ext=buns.priv" (full error message in dump in 
attachment). Anybody any idea?

I also attach the config.log files from thte compilation directory.

Best,
Robin


--enable-pftijah
SuSE 10.2
autoconf-2.60-21
automake-1.9.6-35
libtool-1.5.22-34

----------------------------------------------------------------------

Comment By: Wouter Alink (vzzzbx)
Date: 2007-10-05 19:05

Message:
Logged In: YES 
user_id=621590
Originator: NO

hello robin, 
i am not sure whether it is completely related to this bug report, but a
while ago we added 'standoff annotation' support to monetdb/xquery. you
might want to have a look at it
(http://repos.project.cwi.nl:8888/cwi_repository/docs/II/11/11086B.pdf),
because one of the reasons to add it was the difficulty of having the
query-optimizer recognize the temporal join (interval A overlaps/contains
interval B).
greetings,
wouter

----------------------------------------------------------------------

Comment By: Stefan Manegold (stmane)
Date: 2007-10-05 16:58

Message:
Logged In: YES 
user_id=572415
Originator: NO

Hi Robin,

if I understand your numbers correctly, the 130 documents have a
original/initial total size of about 4MB,
and after all inserts (provided they succeed) the total size should grow
to about 18 MB, right.

Still open: How much physical memory and free disk space (on the partition
you're using for your dbfarm and xquery_logdir) does your machine have?

In any case, I consider the amount of data rather small and would not
expect the reported error.

Hence, could you provided us (via an attached or downloadable .tar.gz)
with all the data (documents), queries, scripts and instructions we need to
reproduce the problem?

Thanks!

Stefan

ps: attaching config.log is good, but (1) it's easier (at least for me) if
I can the important info in the bug report instead of grepping through a
large config.log, and (2) often, (much) more info the available in
config.log is required (see above).


----------------------------------------------------------------------

Comment By: Robin Aly (raly)
Date: 2007-10-05 16:24

Message:
Logged In: YES 
user_id=1423368
Originator: YES

Stefan
thanks for your answer. I configured the MonetDB + Pathfinder as follows:
MonetDB:
/local/alyr/dev/src/MonetDB-2007-09-06/configure --enable-debug
--prefix=/lo
cal/alyr/dev/root/monetdb-2007-09-06

Pathfinder
/local/alyr/dev/src/pathfinder-2007-09-06/configure --enable-debug
--enable-
pftijah --prefix=/local/alyr/dev/root/monetdb-2007-09-06

I have a 32 bit machine (uname -m = i686)

Here are the statistics over the data set:
I achieved the problem using another way so 
I can only give you the size after the desired
insert:

             min max         avg s    sum [bytes]
Orig Files   3965   155987   37455.2  4082618
Result Files 16361  625657   168992   17913114

Yes, It works fine for documents 1-10.

Suggestions: I discussed with Maurice, that it probably would be good
for bug tracking if everybody appends his config.logs to the report.
This way ./configure calls, library versions, architecture etc are all
visible. No critique - only a suggestion. 

----------------------------------------------------------------------

Comment By: Stefan Manegold (stmane)
Date: 2007-10-04 00:46

Message:
Logged In: YES 
user_id=572415
Originator: NO

Robin,

what kind of hardware are you using (CPU type, amount physical memory,
amount of swap, free disk space)?
Are you using a 32-bit or 64-bit CPU/system/OS?
In case it's 64-bit are you using a 32-bit or 64-bit MonetDB/XQuery?
In case of 64-bit MonetDB/XQuery, are you using 64-bit (default) or 32-bit
(--enable-oid32) OIDs?

>From your error message, I see that MonetDB fails to mmap ~487 MB possibly
(on a 32-bit system) because the address space is full/exhausted...

How large are the 130 documents you're using (min, max, total) and how
much data do you (try to) insert (per doc and in total).

Did I get it right that you load all 130 documents before starting the
inserts and the loading is successful?
Inserting (using integrate.xq) works fine for the first 10 documents, but
fails for the 11th?


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1792353&group_id=56967

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs

Reply via email to