hi everyone an attempt to set up a cluster, I'm reading around and see some howto writers would say to put "private dir on the FS cluster" - one question I have: is this correct? necessary? I have partial success, I get: $ ctdb status Number of nodes:2 pnn:0 10.5.6.32 OK pnn:1 10.5.6.49 UNHEALTHY (THIS NODE) Generation:323266562 Size:2 hash:0 lmaster:0 hash:1 lmaster:1 Recovery mode:NORMAL (0) Recovery master:0 I use: passdb backend = ldapsam:"ldapi:///var/run/ldapi" idmap backend = ldap:"ldapi:///var/run/ldapi" and on that one "unhealthy" node I see: 50.samba: ERROR: smb.conf cache create failed would you care to make some sense out of it for me please?
On 09/11/16 16:05, lejeczek via samba wrote:> hi everyone > > an attempt to set up a cluster, I'm reading around and see > some howto writers would say to put "private dir on the FS > cluster" - one question I have: is this correct? necessary? > > I have partial success, I get: > > $ ctdb status > Number of nodes:2 > pnn:0 10.5.6.32 OK > pnn:1 10.5.6.49 UNHEALTHY (THIS NODE) > Generation:323266562 > Size:2 > hash:0 lmaster:0 > hash:1 lmaster:1 > Recovery mode:NORMAL (0) > Recovery master:0 > > I use: > passdb backend = ldapsam:"ldapi:///var/run/ldapi" > idmap backend = ldap:"ldapi:///var/run/ldapi" > > and on that one "unhealthy" node I see: > > 50.samba: ERROR: smb.conf cache create failed >why would this one server be unable to create that file? If I copy it manually while ctdb is running I see: 50.samba: WARNING: smb.conf cache update failed - using old cache file Node became HEALTHY. Ask recovery master 0 to perform ip reallocation and: $ ctdb status Number of nodes:2 pnn:0 10.5.6.32 OK pnn:1 10.5.6.49 OK (THIS NODE) but only so far as next stop/restart. What the heck?> > would you care to make some sense out of it for me please? >
ough.. gee. never mind that (always have your selinux toolkit at hand) On 09/11/16 17:03, lejeczek via samba wrote:> > > On 09/11/16 16:05, lejeczek via samba wrote: >> hi everyone >> >> an attempt to set up a cluster, I'm reading around and >> see some howto writers would say to put "private dir on >> the FS cluster" - one question I have: is this correct? >> necessary? >> >> I have partial success, I get: >> >> $ ctdb status >> Number of nodes:2 >> pnn:0 10.5.6.32 OK >> pnn:1 10.5.6.49 UNHEALTHY (THIS NODE) >> Generation:323266562 >> Size:2 >> hash:0 lmaster:0 >> hash:1 lmaster:1 >> Recovery mode:NORMAL (0) >> Recovery master:0 >> >> I use: >> passdb backend = ldapsam:"ldapi:///var/run/ldapi" >> idmap backend = ldap:"ldapi:///var/run/ldapi" >> >> and on that one "unhealthy" node I see: >> >> 50.samba: ERROR: smb.conf cache create failed >> > why would this one server be unable to create that file? > If I copy it manually while ctdb is running I see: > > 50.samba: WARNING: smb.conf cache update failed - using > old cache file > Node became HEALTHY. Ask recovery master 0 to perform ip > reallocation > > and: > > $ ctdb status > Number of nodes:2 > pnn:0 10.5.6.32 OK > pnn:1 10.5.6.49 OK (THIS NODE) > > but only so far as next stop/restart. What the heck? > >> >> would you care to make some sense out of it for me please? >> > >
On Wed, 9 Nov 2016 16:05:32 +0000, lejeczek via samba <samba at lists.samba.org> wrote:> an attempt to set up a cluster, I'm reading around and see > some howto writers would say to put "private dir on the FS > cluster" - one question I have: is this correct? necessary?No, this is not necessary. Where are you seeing this? If it is in any documentation on samba.org then I'll fix it... Thanks! peace & happiness, martin
On 09/11/16 23:28, Martin Schwenke via samba wrote:> On Wed, 9 Nov 2016 16:05:32 +0000, lejeczek via samba > <samba at lists.samba.org> wrote: > >> an attempt to set up a cluster, I'm reading around and see >> some howto writers would say to put "private dir on the FS >> cluster" - one question I have: is this correct? necessary? > No, this is not necessary. > > Where are you seeing this? If it is in any documentation on samba.org > then I'll fix it...not on samba.org (although have not plough through it), but glusterfs own howtos (a various referring to gluster) say to put it there. https://support.rackspace.com/how-to/glusterfs-high-availability-through-ctdb/ - but cannot see they say why. I my case - ldap is the backend so I'm even more dubious.> > Thanks! > > peace & happiness, > martin >
Apparently Analagous Threads
- CTDB and samba private dir (+ldap)
- CTDB and samba private dir (+ldap)
- CTDB and nfs-ganesha
- glusterfs + ctdb + nfs-ganesha , unplug the network cable of serving node, takes around ~20 mins for IO to resume
- Upgrading a ctdb cluster: samba not listening on TCP port 445