Re: [Archive-access-discuss] Nutchwax+Wera problems

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Jo=E3o Cl=E1udio Luzio wrote:
> Oops.. forgot to say that the arcs where on=20
> /var/local/webarchive/heritrix/jobs/bn_18_test-20060619172505727/=20
> directory but with .arc.gz instead of .arc.
>
>  =20
This should be fine.

> Jo=E3o Cl=E1udio Luzio wrote:
>  =20
>> Hi,
>>     I've been trying to get the pair up and running for a while now bu=
t=20
>> had some problems..
>>     Using nutchwax 0.4.3 and the wera (0.4.2RC1 & 0.5.0) I managed to=20
>> get it running but some of the related files (images)
>> aren't displayed. Those get:
>> <retrievermessage>
>>     <head>
>>         <errorcode>4</errorcode>
>>         <errormessage>Unable to parse Archive Identifier</errormessage=
>
>>     </head>
>> </retrievermessage>
>>     Using wera debug I found that the "[archiveidentifier] =3D>=20
>> 2770/IAH-20060619172903-00000-webarchive1" for a specific search i mad=
e.
>> (Starting tomcat from the nutchwax indexed data)
>>    =20

So, it generally works but some of the images don't show sometimes?

>>     Using wayback I dont have the same problems(I dont use nutchwax wi=
th=20
>> wayback..).
>>
>>     I've tried to get nutchwax 0.6.1 and wera running but the opensear=
ch=20
>> servlet for the rss from nutchwax gives an exception..
>>    =20

Do you still have the exception?

>>     So i tried nutchwax 0.7.0 (with latest hadoop - standalone), but n=
ow=20
>> the arcretriever gives an exception when trying to get the document.
>>     Using wera debug I found that the "[archiveidentifier] =3D>=20
>> 2234331/filedesc://IAH-20060619172903-00000-webarchive1.arc" for the=20
>> same search i made.
>> (Starting tomcat from anywhere)
>>
>> The exception:
>> 7  Bad function argument Cause: java.io.FileNotFoundException:=20
>> /var/local/webarchive/heritrix/jobs/bn_18_test-20060619172505727/filed=
esc:/IAH-20060619172903-00000-webarchive1.arc=20
>> does not exist. Stack trace:=20
>> org.archive.io.arc.ARCUtils.isReadable(ARCUtils.java:171)=20
>> org.archive.io.arc.ARCUtils.testCompressedARCFile(ARCUtils.java:94)=20
>> org.archive.io.arc.ARCReaderFactory.get(ARCReaderFactory.java:200)=20
>> org.archive.io.arc.ARCReaderFactory.get(ARCReaderFactory.java:194)=20
>> no.nb.nwa.retriever.ARCRetriever.getDocument(ARCRetriever.java:410)=20
>> no.nb.nwa.retriever.ARCRetriever.doGet(ARCRetriever.java:131)
>>    =20

Looks like we shouldn't be putting the 'filedesc:' on front of ARC=20
name?  Does ARCRetreiver work if you make a request with=20
IAH-20060619172903-00000-webarchive1.arc instead of=20
filedesc:/IAH-20060619172903-00000-webarchive1.arc?

St.Ack