| Message ID | 1303032724-13456-2-git-send-email-wido@widodh.nl (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi Wido, Thanks for the patch. This was already fixed by e34e332bc43b30f43dee3f2b153ea10977553068. I don't think we want log to syslog to be in the sample configuration file. Maybe we could include it, but commented out. Most users will want to enable logging to a file or to syslog, but not both. Colin On Sun, Apr 17, 2011 at 2:32 AM, Wido den Hollander <wido@widodh.nl> wrote: > > Signed-off-by: Wido den Hollander <wido@widodh.nl> > --- > src/sample.ceph.conf | 23 +++++++++++++---------- > 1 files changed, 13 insertions(+), 10 deletions(-) > > diff --git a/src/sample.ceph.conf b/src/sample.ceph.conf > index bb874e7..31674f6 100644 > --- a/src/sample.ceph.conf > +++ b/src/sample.ceph.conf > @@ -24,6 +24,9 @@ > ; set up pid files > pid file = /var/run/ceph/$name.pid > > + ; Also log to syslog > + log to syslog = true > + > ; monitors > ; You need at least one. You need at least three if you want to > ; tolerate any node failures. Always create an odd number. > @@ -37,15 +40,15 @@ > ;debug paxos = 20 > ;debug auth = 20 > > -[mon0] > +[mon.alpha] > host = alpha > mon addr = 192.168.0.10:6789 > > -[mon1] > +[mon.beta] > host = beta > mon addr = 192.168.0.11:6789 > > -[mon2] > +[mon.gamma] > host = gamma > mon addr = 192.168.0.12:6789 > > @@ -70,16 +73,16 @@ > ; Define as many as you like. > [osd] > ; This is where the btrfs volume will be mounted. > - osd data = /data/osd$id > + osd data = /data/$name > > ; Ideally, make this a separate disk or partition. A few > ; hundred MB should be enough; more if you have fast or many > ; disks. You can use a file under the osd data dir if need be > - ; (e.g. /data/osd$id/journal), but it will be slower than a > + ; (e.g. /data/$name/journal), but it will be slower than a > ; separate disk or partition. > > ; This is an example of a file-based journal. > - osd journal = /data/osd$id/journal > + osd journal = /data/$name/journal > osd journal size = 1000 ; journal size, in megabytes > > ; osd logging to debug osd issues, in order of likelihood of being > @@ -89,7 +92,7 @@ > ;debug filestore = 20 > ;debug journal = 20 > > -[osd0] > +[osd.0] > host = delta > > ; if 'btrfs devs' is not specified, you're responsible for > @@ -98,14 +101,14 @@ > ; usually fine for basic testing). > btrfs devs = /dev/sdx > > -[osd1] > +[osd.1] > host = epsilon > btrfs devs = /dev/sdy > > -[osd2] > +[osd.2] > host = zeta > btrfs devs = /dev/sdx > > -[osd3] > +[osd.3] > host = eta > btrfs devs = /dev/sdy > -- > 1.7.0.4 > > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi, On Sun, 2011-04-17 at 17:37 -0700, Colin McCabe wrote: > I don't think we want log to syslog to be in the sample configuration > file. Maybe we could include it, but commented out. Most users will > want to enable logging to a file or to syslog, but not both. > Ah yes, I see. The syslog was my bad, my idea was to have it commented out, but have it just as a reference in there, didn't check that good enough. But right now you can only disable logging to a file by setting: log file = That seems a bit odd, "log file = false" would be easier. Imho the current sample config is still broken. Monitors names are only numeric, shouldn't we encourage users to give them a full name? mon.alpha, mon.beta, etc? osd journal = /data/osd$id/journal That will result in: /data/osd0/journal When the osd's name is osd.0, that also seems a bit weird, would $name be better there then osd$id (see my patch). Same goes for data btw Wido > Colin > > > On Sun, Apr 17, 2011 at 2:32 AM, Wido den Hollander <wido@widodh.nl> wrote: > > > > Signed-off-by: Wido den Hollander <wido@widodh.nl> > > --- > > src/sample.ceph.conf | 23 +++++++++++++---------- > > 1 files changed, 13 insertions(+), 10 deletions(-) > > > > diff --git a/src/sample.ceph.conf b/src/sample.ceph.conf > > index bb874e7..31674f6 100644 > > --- a/src/sample.ceph.conf > > +++ b/src/sample.ceph.conf > > @@ -24,6 +24,9 @@ > > ; set up pid files > > pid file = /var/run/ceph/$name.pid > > > > + ; Also log to syslog > > + log to syslog = true > > + > > ; monitors > > ; You need at least one. You need at least three if you want to > > ; tolerate any node failures. Always create an odd number. > > @@ -37,15 +40,15 @@ > > ;debug paxos = 20 > > ;debug auth = 20 > > > > -[mon0] > > +[mon.alpha] > > host = alpha > > mon addr = 192.168.0.10:6789 > > > > -[mon1] > > +[mon.beta] > > host = beta > > mon addr = 192.168.0.11:6789 > > > > -[mon2] > > +[mon.gamma] > > host = gamma > > mon addr = 192.168.0.12:6789 > > > > @@ -70,16 +73,16 @@ > > ; Define as many as you like. > > [osd] > > ; This is where the btrfs volume will be mounted. > > - osd data = /data/osd$id > > + osd data = /data/$name > > > > ; Ideally, make this a separate disk or partition. A few > > ; hundred MB should be enough; more if you have fast or many > > ; disks. You can use a file under the osd data dir if need be > > - ; (e.g. /data/osd$id/journal), but it will be slower than a > > + ; (e.g. /data/$name/journal), but it will be slower than a > > ; separate disk or partition. > > > > ; This is an example of a file-based journal. > > - osd journal = /data/osd$id/journal > > + osd journal = /data/$name/journal > > osd journal size = 1000 ; journal size, in megabytes > > > > ; osd logging to debug osd issues, in order of likelihood of being > > @@ -89,7 +92,7 @@ > > ;debug filestore = 20 > > ;debug journal = 20 > > > > -[osd0] > > +[osd.0] > > host = delta > > > > ; if 'btrfs devs' is not specified, you're responsible for > > @@ -98,14 +101,14 @@ > > ; usually fine for basic testing). > > btrfs devs = /dev/sdx > > > > -[osd1] > > +[osd.1] > > host = epsilon > > btrfs devs = /dev/sdy > > > > -[osd2] > > +[osd.2] > > host = zeta > > btrfs devs = /dev/sdx > > > > -[osd3] > > +[osd.3] > > host = eta > > btrfs devs = /dev/sdy > > -- > > 1.7.0.4 > > > > -- > > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, Apr 18, 2011 at 12:01 AM, Wido den Hollander <wido@widodh.nl> wrote: > Hi, > > On Sun, 2011-04-17 at 17:37 -0700, Colin McCabe wrote: >> I don't think we want log to syslog to be in the sample configuration >> file. Maybe we could include it, but commented out. Most users will >> want to enable logging to a file or to syslog, but not both. >> > > Ah yes, I see. The syslog was my bad, my idea was to have it commented > out, but have it just as a reference in there, didn't check that good > enough. > > But right now you can only disable logging to a file by setting: > > log file = > > That seems a bit odd, "log file = false" would be easier. Disabling the log file with an empty string seems pretty intuitive to me. We used to have a separate boolean controlling this, and it was very confusing to users. > > Imho the current sample config is still broken. > > Monitors names are only numeric, shouldn't we encourage users to give > them a full name? mon.alpha, mon.beta, etc? > > osd journal = /data/osd$id/journal > > That will result in: /data/osd0/journal > > When the osd's name is osd.0, that also seems a bit weird, would $name > be better there then osd$id (see my patch). > > Same goes for data btw It would be nice to have all of $name, $type and $id somewhere in there so that users know they exist. Aside from that, I don't really have a strong opinion... cheers, Colin > > Wido > >> Colin >> >> >> On Sun, Apr 17, 2011 at 2:32 AM, Wido den Hollander <wido@widodh.nl> wrote: >> > >> > Signed-off-by: Wido den Hollander <wido@widodh.nl> >> > --- >> > src/sample.ceph.conf | 23 +++++++++++++---------- >> > 1 files changed, 13 insertions(+), 10 deletions(-) >> > >> > diff --git a/src/sample.ceph.conf b/src/sample.ceph.conf >> > index bb874e7..31674f6 100644 >> > --- a/src/sample.ceph.conf >> > +++ b/src/sample.ceph.conf >> > @@ -24,6 +24,9 @@ >> > ; set up pid files >> > pid file = /var/run/ceph/$name.pid >> > >> > + ; Also log to syslog >> > + log to syslog = true >> > + >> > ; monitors >> > ; You need at least one. You need at least three if you want to >> > ; tolerate any node failures. Always create an odd number. >> > @@ -37,15 +40,15 @@ >> > ;debug paxos = 20 >> > ;debug auth = 20 >> > >> > -[mon0] >> > +[mon.alpha] >> > host = alpha >> > mon addr = 192.168.0.10:6789 >> > >> > -[mon1] >> > +[mon.beta] >> > host = beta >> > mon addr = 192.168.0.11:6789 >> > >> > -[mon2] >> > +[mon.gamma] >> > host = gamma >> > mon addr = 192.168.0.12:6789 >> > >> > @@ -70,16 +73,16 @@ >> > ; Define as many as you like. >> > [osd] >> > ; This is where the btrfs volume will be mounted. >> > - osd data = /data/osd$id >> > + osd data = /data/$name >> > >> > ; Ideally, make this a separate disk or partition. A few >> > ; hundred MB should be enough; more if you have fast or many >> > ; disks. You can use a file under the osd data dir if need be >> > - ; (e.g. /data/osd$id/journal), but it will be slower than a >> > + ; (e.g. /data/$name/journal), but it will be slower than a >> > ; separate disk or partition. >> > >> > ; This is an example of a file-based journal. >> > - osd journal = /data/osd$id/journal >> > + osd journal = /data/$name/journal >> > osd journal size = 1000 ; journal size, in megabytes >> > >> > ; osd logging to debug osd issues, in order of likelihood of being >> > @@ -89,7 +92,7 @@ >> > ;debug filestore = 20 >> > ;debug journal = 20 >> > >> > -[osd0] >> > +[osd.0] >> > host = delta >> > >> > ; if 'btrfs devs' is not specified, you're responsible for >> > @@ -98,14 +101,14 @@ >> > ; usually fine for basic testing). >> > btrfs devs = /dev/sdx >> > >> > -[osd1] >> > +[osd.1] >> > host = epsilon >> > btrfs devs = /dev/sdy >> > >> > -[osd2] >> > +[osd.2] >> > host = zeta >> > btrfs devs = /dev/sdx >> > >> > -[osd3] >> > +[osd.3] >> > host = eta >> > btrfs devs = /dev/sdy >> > -- >> > 1.7.0.4 >> > >> > -- >> > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in >> > the body of a message to majordomo@vger.kernel.org >> > More majordomo info at http://vger.kernel.org/majordomo-info.html >> > > > > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, 18 Apr 2011, Colin McCabe wrote: > On Mon, Apr 18, 2011 at 12:01 AM, Wido den Hollander <wido@widodh.nl> wrote: > > Hi, > > > > On Sun, 2011-04-17 at 17:37 -0700, Colin McCabe wrote: > >> I don't think we want log to syslog to be in the sample configuration > >> file. Maybe we could include it, but commented out. Most users will > >> want to enable logging to a file or to syslog, but not both. > >> > > > > Ah yes, I see. The syslog was my bad, my idea was to have it commented > > out, but have it just as a reference in there, didn't check that good > > enough. > > > > But right now you can only disable logging to a file by setting: > > > > log file = > > > > That seems a bit odd, "log file = false" would be easier. > > Disabling the log file with an empty string seems pretty intuitive to > me. We used to have a separate boolean controlling this, and it was > very confusing to users. > > > > > Imho the current sample config is still broken. > > > > Monitors names are only numeric, shouldn't we encourage users to give > > them a full name? mon.alpha, mon.beta, etc? Yeah.. I thought they were a, b, c, but if they're not, they should be! Using numbers for the id is a recipe for confusion because they are dynamically assigned numerical ids at runtime irrespective of their alphanumeric id. > > osd journal = /data/osd$id/journal > > > > That will result in: /data/osd0/journal > > > > When the osd's name is osd.0, that also seems a bit weird, would $name > > be better there then osd$id (see my patch). > > > > Same goes for data btw > > It would be nice to have all of $name, $type and $id somewhere in > there so that users know they exist. Aside from that, I don't really > have a strong opinion... We should be consistent with the . throughout, whatever we do. I'm fine with osd.$id or just $name (e.g. /data/$name)... sage > > cheers, > Colin > > > > > Wido > > > >> Colin > >> > >> > >> On Sun, Apr 17, 2011 at 2:32 AM, Wido den Hollander <wido@widodh.nl> wrote: > >> > > >> > Signed-off-by: Wido den Hollander <wido@widodh.nl> > >> > --- > >> > src/sample.ceph.conf | 23 +++++++++++++---------- > >> > 1 files changed, 13 insertions(+), 10 deletions(-) > >> > > >> > diff --git a/src/sample.ceph.conf b/src/sample.ceph.conf > >> > index bb874e7..31674f6 100644 > >> > --- a/src/sample.ceph.conf > >> > +++ b/src/sample.ceph.conf > >> > @@ -24,6 +24,9 @@ > >> > ; set up pid files > >> > pid file = /var/run/ceph/$name.pid > >> > > >> > + ; Also log to syslog > >> > + log to syslog = true > >> > + > >> > ; monitors > >> > ; You need at least one. You need at least three if you want to > >> > ; tolerate any node failures. Always create an odd number. > >> > @@ -37,15 +40,15 @@ > >> > ;debug paxos = 20 > >> > ;debug auth = 20 > >> > > >> > -[mon0] > >> > +[mon.alpha] > >> > host = alpha > >> > mon addr = 192.168.0.10:6789 > >> > > >> > -[mon1] > >> > +[mon.beta] > >> > host = beta > >> > mon addr = 192.168.0.11:6789 > >> > > >> > -[mon2] > >> > +[mon.gamma] > >> > host = gamma > >> > mon addr = 192.168.0.12:6789 > >> > > >> > @@ -70,16 +73,16 @@ > >> > ; Define as many as you like. > >> > [osd] > >> > ; This is where the btrfs volume will be mounted. > >> > - osd data = /data/osd$id > >> > + osd data = /data/$name > >> > > >> > ; Ideally, make this a separate disk or partition. A few > >> > ; hundred MB should be enough; more if you have fast or many > >> > ; disks. You can use a file under the osd data dir if need be > >> > - ; (e.g. /data/osd$id/journal), but it will be slower than a > >> > + ; (e.g. /data/$name/journal), but it will be slower than a > >> > ; separate disk or partition. > >> > > >> > ; This is an example of a file-based journal. > >> > - osd journal = /data/osd$id/journal > >> > + osd journal = /data/$name/journal > >> > osd journal size = 1000 ; journal size, in megabytes > >> > > >> > ; osd logging to debug osd issues, in order of likelihood of being > >> > @@ -89,7 +92,7 @@ > >> > ;debug filestore = 20 > >> > ;debug journal = 20 > >> > > >> > -[osd0] > >> > +[osd.0] > >> > host = delta > >> > > >> > ; if 'btrfs devs' is not specified, you're responsible for > >> > @@ -98,14 +101,14 @@ > >> > ; usually fine for basic testing). > >> > btrfs devs = /dev/sdx > >> > > >> > -[osd1] > >> > +[osd.1] > >> > host = epsilon > >> > btrfs devs = /dev/sdy > >> > > >> > -[osd2] > >> > +[osd.2] > >> > host = zeta > >> > btrfs devs = /dev/sdx > >> > > >> > -[osd3] > >> > +[osd.3] > >> > host = eta > >> > btrfs devs = /dev/sdy > >> > -- > >> > 1.7.0.4 > >> > > >> > -- > >> > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > >> > the body of a message to majordomo@vger.kernel.org > >> > More majordomo info at http://vger.kernel.org/majordomo-info.html > >> > > > > > > > > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > >
diff --git a/src/sample.ceph.conf b/src/sample.ceph.conf index bb874e7..31674f6 100644 --- a/src/sample.ceph.conf +++ b/src/sample.ceph.conf @@ -24,6 +24,9 @@ ; set up pid files pid file = /var/run/ceph/$name.pid + ; Also log to syslog + log to syslog = true + ; monitors ; You need at least one. You need at least three if you want to ; tolerate any node failures. Always create an odd number. @@ -37,15 +40,15 @@ ;debug paxos = 20 ;debug auth = 20 -[mon0] +[mon.alpha] host = alpha mon addr = 192.168.0.10:6789 -[mon1] +[mon.beta] host = beta mon addr = 192.168.0.11:6789 -[mon2] +[mon.gamma] host = gamma mon addr = 192.168.0.12:6789 @@ -70,16 +73,16 @@ ; Define as many as you like. [osd] ; This is where the btrfs volume will be mounted. - osd data = /data/osd$id + osd data = /data/$name ; Ideally, make this a separate disk or partition. A few ; hundred MB should be enough; more if you have fast or many ; disks. You can use a file under the osd data dir if need be - ; (e.g. /data/osd$id/journal), but it will be slower than a + ; (e.g. /data/$name/journal), but it will be slower than a ; separate disk or partition. ; This is an example of a file-based journal. - osd journal = /data/osd$id/journal + osd journal = /data/$name/journal osd journal size = 1000 ; journal size, in megabytes ; osd logging to debug osd issues, in order of likelihood of being @@ -89,7 +92,7 @@ ;debug filestore = 20 ;debug journal = 20 -[osd0] +[osd.0] host = delta ; if 'btrfs devs' is not specified, you're responsible for @@ -98,14 +101,14 @@ ; usually fine for basic testing). btrfs devs = /dev/sdx -[osd1] +[osd.1] host = epsilon btrfs devs = /dev/sdy -[osd2] +[osd.2] host = zeta btrfs devs = /dev/sdx -[osd3] +[osd.3] host = eta btrfs devs = /dev/sdy
Signed-off-by: Wido den Hollander <wido@widodh.nl> --- src/sample.ceph.conf | 23 +++++++++++++---------- 1 files changed, 13 insertions(+), 10 deletions(-)