> Indexes Backup

["Andrea K." <ak...@alteanet.it>]

Hi all,
Can you help me to find the right solution to backup a clustered JR?

Details are:
1. JR on Oracle 10g database (repository and cluster tables) - NO BACKUP
PROBLEMS
2. Local directories (each server) with configuration files and indexes (+
revision.log)

How can I backup it to take the situation coherent to db and cluster nodes
view?

A question is (I describe a situation):
- Server 1 is updated to revision 120 (i.e.) and I backup it (server is
stopped).
- Server 2 is updated in the meanwhile to 124 (i.e.) and I backup it (server
is stopped).
- Server 1 is restarted and write some revisions up to 127.
- Server 1 is restarted and write some revisions up to 135.

If Server 1 crash and i re-apply backup indexes up to 120, will it work or
some revisions are skipped during Server1 initial reindex as generator is
itself?

Thanks a lot for your help.
BR,
Andrea -
-- 
View this message in context: http://www.nabble.com/%3E-Indexes-Backup-tf4805240.html#a13747251
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.

Re: > Indexes Backup

["Andrea K." <ak...@alteanet.it>]

So, regarding my starting example.
If I restore indexes to revision 120 on Server1, some future (self) revision
will be skipped?

In this case, there's no way (or reason) to backup indexes!

You can never return to a consistent index because some revisions will be
skipped by Server1, and consequently, some data will never be searcheable on
Server1. Is it right or I'm wrong?

Is it a correct behaviour also if the server restarts from an old revision?!

So, any idea to backup the system/indexes in a correct way?

Thanks...
Andrea.



Dominique Pfister wrote:
> 
> On 19/11/2007, Marcel Reutegger <ma...@gmx.net> wrote:
>> Andrea K. wrote:
>> > Hi Marcel,
>> > my question is exactly over this point.
>> >
>> > I know that it will not re-index all the repository. As I often see
>> messages
>> > in repository log referring to "revision added by local and skipped" -
>> my
>> > specif question is: these revisions are really skipped in a case like
>> this
>> > or restarting from a minor revision it will reload them all? In other
>> works,
>> > Server1 will read and reload all self Server1 revisions or skip them
>> and
>> > read only Server2 revisions?
>>
>> that's also my understanding of how clustering works. but maybe dominique
>> can
>> jump in and correct me if I'm wrong here.
>>
> 
> This is exactly what will happen: a cluster node will always ignore
> entries it has made itself.
> 
> Kind regards
> Dominique
> 
> 

-- 
View this message in context: http://www.nabble.com/%3E-Indexes-Backup-tf4805240.html#a13830816
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.

Re: > Indexes Backup

[Dominique Pfister <do...@day.com>]

On 19/11/2007, Marcel Reutegger <ma...@gmx.net> wrote:
> Andrea K. wrote:
> > Hi Marcel,
> > my question is exactly over this point.
> >
> > I know that it will not re-index all the repository. As I often see messages
> > in repository log referring to "revision added by local and skipped" - my
> > specif question is: these revisions are really skipped in a case like this
> > or restarting from a minor revision it will reload them all? In other works,
> > Server1 will read and reload all self Server1 revisions or skip them and
> > read only Server2 revisions?
>
> that's also my understanding of how clustering works. but maybe dominique can
> jump in and correct me if I'm wrong here.
>

This is exactly what will happen: a cluster node will always ignore
entries it has made itself.

Kind regards
Dominique

Re: > Indexes Backup

[Marcel Reutegger <ma...@gmx.net>]

Andrea K. wrote:
> Hi Marcel,
> my question is exactly over this point.
> 
> I know that it will not re-index all the repository. As I often see messages
> in repository log referring to "revision added by local and skipped" - my
> specif question is: these revisions are really skipped in a case like this
> or restarting from a minor revision it will reload them all? In other works,
> Server1 will read and reload all self Server1 revisions or skip them and
> read only Server2 revisions?

that's also my understanding of how clustering works. but maybe dominique can 
jump in and correct me if I'm wrong here.

I'm not sure what the log message is about. but looking at the code it seems you 
are referring to AbstractJournal.doSync(), is that correct?

regards
  marcel

Re: > Indexes Backup

["Andrea K." <ak...@alteanet.it>]

Hi Marcel,
my question is exactly over this point.

I know that it will not re-index all the repository. As I often see messages
in repository log referring to "revision added by local and skipped" - my
specif question is: these revisions are really skipped in a case like this
or restarting from a minor revision it will reload them all? In other works,
Server1 will read and reload all self Server1 revisions or skip them and
read only Server2 revisions?

Thanks a lot.

BR,
Andrea -



Marcel Reutegger wrote:
> 
> Hi Andrea,
> 
> if the backup of server 1 includes the index files, then after a restore
> of the 
> files and a restart of the server jackrabbit will not re-index content
> because 
> the index is already present. instead jackrabbit on server 1 will trigger
> events 
> for the revisions that it missed while it was shutdown. this will also
> update 
> the index to the latest revision available on the database.
> 
> regards
>   marcel
> 
> Andrea K. wrote:
>> Hi all,
>> Can you help me to find the right solution to backup a clustered JR?
>> 
>> Details are:
>> 1. JR on Oracle 10g database (repository and cluster tables) - NO BACKUP
>> PROBLEMS
>> 2. Local directories (each server) with configuration files and indexes
>> (+
>> revision.log)
>> 
>> How can I backup it to take the situation coherent to db and cluster
>> nodes
>> view?
>> 
>> A question is (I describe a situation):
>> - Server 1 is updated to revision 120 (i.e.) and I backup it (server is
>> stopped).
>> - Server 2 is updated in the meanwhile to 124 (i.e.) and I backup it
>> (server
>> is stopped).
>> - Server 1 is restarted and write some revisions up to 127.
>> - Server 1 is restarted and write some revisions up to 135.
>> 
>> If Server 1 crash and i re-apply backup indexes up to 120, will it work
>> or
>> some revisions are skipped during Server1 initial reindex as generator is
>> itself?
>> 
>> Thanks a lot for your help.
>> BR,
>> Andrea -
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/%3E-Indexes-Backup-tf4805240.html#a13830059
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.

Re: > Indexes Backup

[Marcel Reutegger <ma...@gmx.net>]

Hi Andrea,

if the backup of server 1 includes the index files, then after a restore of the 
files and a restart of the server jackrabbit will not re-index content because 
the index is already present. instead jackrabbit on server 1 will trigger events 
for the revisions that it missed while it was shutdown. this will also update 
the index to the latest revision available on the database.

regards
  marcel

Andrea K. wrote:
> Hi all,
> Can you help me to find the right solution to backup a clustered JR?
> 
> Details are:
> 1. JR on Oracle 10g database (repository and cluster tables) - NO BACKUP
> PROBLEMS
> 2. Local directories (each server) with configuration files and indexes (+
> revision.log)
> 
> How can I backup it to take the situation coherent to db and cluster nodes
> view?
> 
> A question is (I describe a situation):
> - Server 1 is updated to revision 120 (i.e.) and I backup it (server is
> stopped).
> - Server 2 is updated in the meanwhile to 124 (i.e.) and I backup it (server
> is stopped).
> - Server 1 is restarted and write some revisions up to 127.
> - Server 1 is restarted and write some revisions up to 135.
> 
> If Server 1 crash and i re-apply backup indexes up to 120, will it work or
> some revisions are skipped during Server1 initial reindex as generator is
> itself?
> 
> Thanks a lot for your help.
> BR,
> Andrea -