“solr.data.dir” can only config a single directory

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|

“solr.data.dir” can only config a single directory

zhenyuan wei
Hi all,
     I found the  “solr.data.dir” can only config a single directory.  I
think it is necessary to be config  multi dirs,such as
”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one disk overload
or capacity limitation.  Any reason to support why not do so?


Best,
TinsWzy
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

Shawn Heisey-2
On 8/27/2018 8:29 PM, zhenyuan wei wrote:
>       I found the  “solr.data.dir” can only config a single directory.  I
> think it is necessary to be config  multi dirs,such as
> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one disk overload
> or capacity limitation.  Any reason to support why not do so?

Nobody has written the code to support it.  It would very likely not be
easy code to write.  Supporting one directory for that setting is pretty
easy ... it would require changing a LOT of existing code to support
more than one.

Thanks,
Shawn

Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

Christopher Schultz
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Shawn,

On 8/27/18 22:37, Shawn Heisey wrote:

> On 8/27/2018 8:29 PM, zhenyuan wei wrote:
>> I found the  “solr.data.dir” can only config a single directory.
>> I think it is necessary to be config  multi dirs,such as
>> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
>> disk overload or capacity limitation.  Any reason to support why
>> not do so?
>
> Nobody has written the code to support it.  It would very likely
> not be easy code to write.  Supporting one directory for that
> setting is pretty easy ... it would require changing a LOT of
> existing code to support more than one.

Also, there are better ways to do this:

- - multi-node Solr with sharding
- - LVM or similar with multi-disk volumes
- - ZFS surely has something for this
- - buy a bigger disk (disk is cheap!)
- - etc.

- -chris
-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
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=rbrT
-----END PGP SIGNATURE-----
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

zhenyuan wei
In reply to this post by Shawn Heisey-2
Is there any issue exists already?my all solr instances (30 or more in the
future ) maybe using this feature because all host will be 4 or more
mounted disks.
I am interesting to watching this feature, and make a contribution to that
 in some way  if possiable.

Shawn Heisey <[hidden email]> 于2018年8月28日周二 上午10:38写道:

> On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> >       I found the  “solr.data.dir” can only config a single directory.  I
> > think it is necessary to be config  multi dirs,such as
> > ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one disk
> overload
> > or capacity limitation.  Any reason to support why not do so?
>
> Nobody has written the code to support it.  It would very likely not be
> easy code to write.  Supporting one directory for that setting is pretty
> easy ... it would require changing a LOT of existing code to support
> more than one.
>
> Thanks,
> Shawn
>
>
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

zhenyuan wei
In reply to this post by Christopher Schultz
@Christopher Schultz
So  you mean that  one 4TB disk is the same as  four  1TB disks ?
HDFS、cassandra、ES.... can do so, multi data path maybe maximize indexing
throughput   in same cases.
click links
<https://www.elastic.co/blog/elasticsearch-performance-indexing-2-0> with
some explain


Christopher Schultz <[hidden email]> 于2018年8月28日周二 上午11:16写道:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> Shawn,
>
> On 8/27/18 22:37, Shawn Heisey wrote:
> > On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> >> I found the  “solr.data.dir” can only config a single directory.
> >> I think it is necessary to be config  multi dirs,such as
> >> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
> >> disk overload or capacity limitation.  Any reason to support why
> >> not do so?
> >
> > Nobody has written the code to support it.  It would very likely
> > not be easy code to write.  Supporting one directory for that
> > setting is pretty easy ... it would require changing a LOT of
> > existing code to support more than one.
>
> Also, there are better ways to do this:
>
> - - multi-node Solr with sharding
> - - LVM or similar with multi-disk volumes
> - - ZFS surely has something for this
> - - buy a bigger disk (disk is cheap!)
> - - etc.
>
> - -chris
> -----BEGIN PGP SIGNATURE-----
> Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
>
> iQIzBAEBCAAdFiEEMmKgYcQvxMe7tcJcHPApP6U8pFgFAluEvn8ACgkQHPApP6U8
> pFgTTg//ayed4AXtocVrB6e/ZK0eWz5/E1Q7Oa7kF0c34l0MH6BIe4iOHDmrR+J9
> A+t6SzVQqURMrDE8plg/xbPTlyGF8wGrEjZUZF4fpWlgnY/qNYxl5S9zJ1hPgBh7
> fCKkb+LuLGdZMM4oORfCYtMgpDjOnLihHmDTfkrvZzyZwOQGeFpgEZDZKFYAjcur
> wqIGTMTTWfSCoPQgQzvI8Husq7Rs75BEc+mAkaPOL0LvT9PQDEPEXXt3Kf5vXgM+
> Eet1ymltZM/Xz+V/em/oeumCoCE18uxi9seuDhTpHRLjS9tCBbPWA0NmobriY3ct
> GskwCnsFDAeGjG/7dcA/zmB8BK4t6JpUvI+OcJU5dvQczpQbhB9WT4GQUiME9Tvr
> RjBES53HoEEKA8gb0kiuPN1pE2MSX8vO3uKpQtzVS2MOmuOeV/IebrnP/zLTll18
> awtWWbPmzaAGAUfXL2ExK0+ism0o31i46CNfLfBBM8jh3lkc2HNdz5TLe8YfN3Sp
> Tj0HfmYynhtH1CggOAcI1M4PIEbIGfoywX/ICSGHnLwfQoDUnBmjqXhGkFUIstWk
> Dcntx+4E4NRny6zDZfg5UMjWYyo+fOVSoaDf6dfgBWIB1I3xPn5Dt0In7+oRtZ9i
> Xlkw6DSaSZZ5caBqjaF278xj7IwEw2zipLPWB7hVCcUhKuJBbDY=
> =rbrT
> -----END PGP SIGNATURE-----
>
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

Erick Erickson
Every _replica_ can point to a different disk. When you do an
ADDREPLICA, then you can supply whatever path to the data
you desire. And you can have as many replicas per Solr instance
as makes sense.

Best,
Erick
On Mon, Aug 27, 2018 at 8:48 PM zhenyuan wei <[hidden email]> wrote:

>
> @Christopher Schultz
> So  you mean that  one 4TB disk is the same as  four  1TB disks ?
> HDFS、cassandra、ES.... can do so, multi data path maybe maximize indexing
> throughput   in same cases.
> click links
> <https://www.elastic.co/blog/elasticsearch-performance-indexing-2-0> with
> some explain
>
>
> Christopher Schultz <[hidden email]> 于2018年8月28日周二 上午11:16写道:
>
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA256
> >
> > Shawn,
> >
> > On 8/27/18 22:37, Shawn Heisey wrote:
> > > On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> > >> I found the  “solr.data.dir” can only config a single directory.
> > >> I think it is necessary to be config  multi dirs,such as
> > >> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
> > >> disk overload or capacity limitation.  Any reason to support why
> > >> not do so?
> > >
> > > Nobody has written the code to support it.  It would very likely
> > > not be easy code to write.  Supporting one directory for that
> > > setting is pretty easy ... it would require changing a LOT of
> > > existing code to support more than one.
> >
> > Also, there are better ways to do this:
> >
> > - - multi-node Solr with sharding
> > - - LVM or similar with multi-disk volumes
> > - - ZFS surely has something for this
> > - - buy a bigger disk (disk is cheap!)
> > - - etc.
> >
> > - -chris
> > -----BEGIN PGP SIGNATURE-----
> > Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
> >
> > iQIzBAEBCAAdFiEEMmKgYcQvxMe7tcJcHPApP6U8pFgFAluEvn8ACgkQHPApP6U8
> > pFgTTg//ayed4AXtocVrB6e/ZK0eWz5/E1Q7Oa7kF0c34l0MH6BIe4iOHDmrR+J9
> > A+t6SzVQqURMrDE8plg/xbPTlyGF8wGrEjZUZF4fpWlgnY/qNYxl5S9zJ1hPgBh7
> > fCKkb+LuLGdZMM4oORfCYtMgpDjOnLihHmDTfkrvZzyZwOQGeFpgEZDZKFYAjcur
> > wqIGTMTTWfSCoPQgQzvI8Husq7Rs75BEc+mAkaPOL0LvT9PQDEPEXXt3Kf5vXgM+
> > Eet1ymltZM/Xz+V/em/oeumCoCE18uxi9seuDhTpHRLjS9tCBbPWA0NmobriY3ct
> > GskwCnsFDAeGjG/7dcA/zmB8BK4t6JpUvI+OcJU5dvQczpQbhB9WT4GQUiME9Tvr
> > RjBES53HoEEKA8gb0kiuPN1pE2MSX8vO3uKpQtzVS2MOmuOeV/IebrnP/zLTll18
> > awtWWbPmzaAGAUfXL2ExK0+ism0o31i46CNfLfBBM8jh3lkc2HNdz5TLe8YfN3Sp
> > Tj0HfmYynhtH1CggOAcI1M4PIEbIGfoywX/ICSGHnLwfQoDUnBmjqXhGkFUIstWk
> > Dcntx+4E4NRny6zDZfg5UMjWYyo+fOVSoaDf6dfgBWIB1I3xPn5Dt0In7+oRtZ9i
> > Xlkw6DSaSZZ5caBqjaF278xj7IwEw2zipLPWB7hVCcUhKuJBbDY=
> > =rbrT
> > -----END PGP SIGNATURE-----
> >
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

zhenyuan wei
But this is not a common way to do so, I mean, nobody want to ADDREPLICA
after collection was created.

Erick Erickson <[hidden email]> 于2018年8月28日周二 下午1:24写道:

> Every _replica_ can point to a different disk. When you do an
> ADDREPLICA, then you can supply whatever path to the data
> you desire. And you can have as many replicas per Solr instance
> as makes sense.
>
> Best,
> Erick
> On Mon, Aug 27, 2018 at 8:48 PM zhenyuan wei <[hidden email]> wrote:
> >
> > @Christopher Schultz
> > So  you mean that  one 4TB disk is the same as  four  1TB disks ?
> > HDFS、cassandra、ES.... can do so, multi data path maybe maximize indexing
> > throughput   in same cases.
> > click links
> > <https://www.elastic.co/blog/elasticsearch-performance-indexing-2-0>
> with
> > some explain
> >
> >
> > Christopher Schultz <[hidden email]> 于2018年8月28日周二
> 上午11:16写道:
> >
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA256
> > >
> > > Shawn,
> > >
> > > On 8/27/18 22:37, Shawn Heisey wrote:
> > > > On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> > > >> I found the  “solr.data.dir” can only config a single directory.
> > > >> I think it is necessary to be config  multi dirs,such as
> > > >> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
> > > >> disk overload or capacity limitation.  Any reason to support why
> > > >> not do so?
> > > >
> > > > Nobody has written the code to support it.  It would very likely
> > > > not be easy code to write.  Supporting one directory for that
> > > > setting is pretty easy ... it would require changing a LOT of
> > > > existing code to support more than one.
> > >
> > > Also, there are better ways to do this:
> > >
> > > - - multi-node Solr with sharding
> > > - - LVM or similar with multi-disk volumes
> > > - - ZFS surely has something for this
> > > - - buy a bigger disk (disk is cheap!)
> > > - - etc.
> > >
> > > - -chris
> > > -----BEGIN PGP SIGNATURE-----
> > > Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
> > >
> > > iQIzBAEBCAAdFiEEMmKgYcQvxMe7tcJcHPApP6U8pFgFAluEvn8ACgkQHPApP6U8
> > > pFgTTg//ayed4AXtocVrB6e/ZK0eWz5/E1Q7Oa7kF0c34l0MH6BIe4iOHDmrR+J9
> > > A+t6SzVQqURMrDE8plg/xbPTlyGF8wGrEjZUZF4fpWlgnY/qNYxl5S9zJ1hPgBh7
> > > fCKkb+LuLGdZMM4oORfCYtMgpDjOnLihHmDTfkrvZzyZwOQGeFpgEZDZKFYAjcur
> > > wqIGTMTTWfSCoPQgQzvI8Husq7Rs75BEc+mAkaPOL0LvT9PQDEPEXXt3Kf5vXgM+
> > > Eet1ymltZM/Xz+V/em/oeumCoCE18uxi9seuDhTpHRLjS9tCBbPWA0NmobriY3ct
> > > GskwCnsFDAeGjG/7dcA/zmB8BK4t6JpUvI+OcJU5dvQczpQbhB9WT4GQUiME9Tvr
> > > RjBES53HoEEKA8gb0kiuPN1pE2MSX8vO3uKpQtzVS2MOmuOeV/IebrnP/zLTll18
> > > awtWWbPmzaAGAUfXL2ExK0+ism0o31i46CNfLfBBM8jh3lkc2HNdz5TLe8YfN3Sp
> > > Tj0HfmYynhtH1CggOAcI1M4PIEbIGfoywX/ICSGHnLwfQoDUnBmjqXhGkFUIstWk
> > > Dcntx+4E4NRny6zDZfg5UMjWYyo+fOVSoaDf6dfgBWIB1I3xPn5Dt0In7+oRtZ9i
> > > Xlkw6DSaSZZ5caBqjaF278xj7IwEw2zipLPWB7hVCcUhKuJBbDY=
> > > =rbrT
> > > -----END PGP SIGNATURE-----
> > >
>
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

Erick Erickson
Patches welcome.

On Mon, Aug 27, 2018, 23:03 zhenyuan wei <[hidden email]> wrote:

> But this is not a common way to do so, I mean, nobody want to ADDREPLICA
> after collection was created.
>
> Erick Erickson <[hidden email]> 于2018年8月28日周二 下午1:24写道:
>
> > Every _replica_ can point to a different disk. When you do an
> > ADDREPLICA, then you can supply whatever path to the data
> > you desire. And you can have as many replicas per Solr instance
> > as makes sense.
> >
> > Best,
> > Erick
> > On Mon, Aug 27, 2018 at 8:48 PM zhenyuan wei <[hidden email]> wrote:
> > >
> > > @Christopher Schultz
> > > So  you mean that  one 4TB disk is the same as  four  1TB disks ?
> > > HDFS、cassandra、ES.... can do so, multi data path maybe maximize
> indexing
> > > throughput   in same cases.
> > > click links
> > > <https://www.elastic.co/blog/elasticsearch-performance-indexing-2-0>
> > with
> > > some explain
> > >
> > >
> > > Christopher Schultz <[hidden email]> 于2018年8月28日周二
> > 上午11:16写道:
> > >
> > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > Hash: SHA256
> > > >
> > > > Shawn,
> > > >
> > > > On 8/27/18 22:37, Shawn Heisey wrote:
> > > > > On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> > > > >> I found the  “solr.data.dir” can only config a single directory.
> > > > >> I think it is necessary to be config  multi dirs,such as
> > > > >> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
> > > > >> disk overload or capacity limitation.  Any reason to support why
> > > > >> not do so?
> > > > >
> > > > > Nobody has written the code to support it.  It would very likely
> > > > > not be easy code to write.  Supporting one directory for that
> > > > > setting is pretty easy ... it would require changing a LOT of
> > > > > existing code to support more than one.
> > > >
> > > > Also, there are better ways to do this:
> > > >
> > > > - - multi-node Solr with sharding
> > > > - - LVM or similar with multi-disk volumes
> > > > - - ZFS surely has something for this
> > > > - - buy a bigger disk (disk is cheap!)
> > > > - - etc.
> > > >
> > > > - -chris
> > > > -----BEGIN PGP SIGNATURE-----
> > > > Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
> > > >
> > > > iQIzBAEBCAAdFiEEMmKgYcQvxMe7tcJcHPApP6U8pFgFAluEvn8ACgkQHPApP6U8
> > > > pFgTTg//ayed4AXtocVrB6e/ZK0eWz5/E1Q7Oa7kF0c34l0MH6BIe4iOHDmrR+J9
> > > > A+t6SzVQqURMrDE8plg/xbPTlyGF8wGrEjZUZF4fpWlgnY/qNYxl5S9zJ1hPgBh7
> > > > fCKkb+LuLGdZMM4oORfCYtMgpDjOnLihHmDTfkrvZzyZwOQGeFpgEZDZKFYAjcur
> > > > wqIGTMTTWfSCoPQgQzvI8Husq7Rs75BEc+mAkaPOL0LvT9PQDEPEXXt3Kf5vXgM+
> > > > Eet1ymltZM/Xz+V/em/oeumCoCE18uxi9seuDhTpHRLjS9tCBbPWA0NmobriY3ct
> > > > GskwCnsFDAeGjG/7dcA/zmB8BK4t6JpUvI+OcJU5dvQczpQbhB9WT4GQUiME9Tvr
> > > > RjBES53HoEEKA8gb0kiuPN1pE2MSX8vO3uKpQtzVS2MOmuOeV/IebrnP/zLTll18
> > > > awtWWbPmzaAGAUfXL2ExK0+ism0o31i46CNfLfBBM8jh3lkc2HNdz5TLe8YfN3Sp
> > > > Tj0HfmYynhtH1CggOAcI1M4PIEbIGfoywX/ICSGHnLwfQoDUnBmjqXhGkFUIstWk
> > > > Dcntx+4E4NRny6zDZfg5UMjWYyo+fOVSoaDf6dfgBWIB1I3xPn5Dt0In7+oRtZ9i
> > > > Xlkw6DSaSZZ5caBqjaF278xj7IwEw2zipLPWB7hVCcUhKuJBbDY=
> > > > =rbrT
> > > > -----END PGP SIGNATURE-----
> > > >
> >
>
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

zhenyuan wei
Pretty cool,here creates an issue to  put this discussion into practice.
issues: https://issues.apache.org/jira/browse/SOLR-12713

Best,
TinsWzy

Erick Erickson <[hidden email]> 于2018年8月28日周二 下午11:51写道:

> Patches welcome.
>
> On Mon, Aug 27, 2018, 23:03 zhenyuan wei <[hidden email]> wrote:
>
> > But this is not a common way to do so, I mean, nobody want to ADDREPLICA
> > after collection was created.
> >
> > Erick Erickson <[hidden email]> 于2018年8月28日周二 下午1:24写道:
> >
> > > Every _replica_ can point to a different disk. When you do an
> > > ADDREPLICA, then you can supply whatever path to the data
> > > you desire. And you can have as many replicas per Solr instance
> > > as makes sense.
> > >
> > > Best,
> > > Erick
> > > On Mon, Aug 27, 2018 at 8:48 PM zhenyuan wei <[hidden email]>
> wrote:
> > > >
> > > > @Christopher Schultz
> > > > So  you mean that  one 4TB disk is the same as  four  1TB disks ?
> > > > HDFS、cassandra、ES.... can do so, multi data path maybe maximize
> > indexing
> > > > throughput   in same cases.
> > > > click links
> > > > <https://www.elastic.co/blog/elasticsearch-performance-indexing-2-0>
> > > with
> > > > some explain
> > > >
> > > >
> > > > Christopher Schultz <[hidden email]> 于2018年8月28日周二
> > > 上午11:16写道:
> > > >
> > > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > > Hash: SHA256
> > > > >
> > > > > Shawn,
> > > > >
> > > > > On 8/27/18 22:37, Shawn Heisey wrote:
> > > > > > On 8/27/2018 8:29 PM, zhenyuan wei wrote:
> > > > > >> I found the  “solr.data.dir” can only config a single directory.
> > > > > >> I think it is necessary to be config  multi dirs,such as
> > > > > >> ”solr.data.dir:/mnt/disk1,/mnt/disk2,/mnt/disk3" , due to one
> > > > > >> disk overload or capacity limitation.  Any reason to support why
> > > > > >> not do so?
> > > > > >
> > > > > > Nobody has written the code to support it.  It would very likely
> > > > > > not be easy code to write.  Supporting one directory for that
> > > > > > setting is pretty easy ... it would require changing a LOT of
> > > > > > existing code to support more than one.
> > > > >
> > > > > Also, there are better ways to do this:
> > > > >
> > > > > - - multi-node Solr with sharding
> > > > > - - LVM or similar with multi-disk volumes
> > > > > - - ZFS surely has something for this
> > > > > - - buy a bigger disk (disk is cheap!)
> > > > > - - etc.
> > > > >
> > > > > - -chris
> > > > > -----BEGIN PGP SIGNATURE-----
> > > > > Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/
> > > > >
> > > > > iQIzBAEBCAAdFiEEMmKgYcQvxMe7tcJcHPApP6U8pFgFAluEvn8ACgkQHPApP6U8
> > > > > pFgTTg//ayed4AXtocVrB6e/ZK0eWz5/E1Q7Oa7kF0c34l0MH6BIe4iOHDmrR+J9
> > > > > A+t6SzVQqURMrDE8plg/xbPTlyGF8wGrEjZUZF4fpWlgnY/qNYxl5S9zJ1hPgBh7
> > > > > fCKkb+LuLGdZMM4oORfCYtMgpDjOnLihHmDTfkrvZzyZwOQGeFpgEZDZKFYAjcur
> > > > > wqIGTMTTWfSCoPQgQzvI8Husq7Rs75BEc+mAkaPOL0LvT9PQDEPEXXt3Kf5vXgM+
> > > > > Eet1ymltZM/Xz+V/em/oeumCoCE18uxi9seuDhTpHRLjS9tCBbPWA0NmobriY3ct
> > > > > GskwCnsFDAeGjG/7dcA/zmB8BK4t6JpUvI+OcJU5dvQczpQbhB9WT4GQUiME9Tvr
> > > > > RjBES53HoEEKA8gb0kiuPN1pE2MSX8vO3uKpQtzVS2MOmuOeV/IebrnP/zLTll18
> > > > > awtWWbPmzaAGAUfXL2ExK0+ism0o31i46CNfLfBBM8jh3lkc2HNdz5TLe8YfN3Sp
> > > > > Tj0HfmYynhtH1CggOAcI1M4PIEbIGfoywX/ICSGHnLwfQoDUnBmjqXhGkFUIstWk
> > > > > Dcntx+4E4NRny6zDZfg5UMjWYyo+fOVSoaDf6dfgBWIB1I3xPn5Dt0In7+oRtZ9i
> > > > > Xlkw6DSaSZZ5caBqjaF278xj7IwEw2zipLPWB7hVCcUhKuJBbDY=
> > > > > =rbrT
> > > > > -----END PGP SIGNATURE-----
> > > > >
> > >
> >
>
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

Bram Van Dam
In reply to this post by zhenyuan wei
On 28/08/18 08:03, zhenyuan wei wrote:
> But this is not a common way to do so, I mean, nobody want to ADDREPLICA
> after collection was created.

I wouldn't say "nobody"..
Reply | Threaded
Open this post in threaded view
|

Re: “solr.data.dir” can only config a single directory

zhenyuan wei
Oh ~   my fault!Sorry for that, I should say somebody,like me~

Bram Van Dam <[hidden email]> 于2018年8月29日周三 下午3:28写道:

> On 28/08/18 08:03, zhenyuan wei wrote:
> > But this is not a common way to do so, I mean, nobody want to ADDREPLICA
> > after collection was created.
>
> I wouldn't say "nobody"..
>