> -----Original Message----- > From: Tim Beale [mailto:timbeale at catalyst.net.nz] > Sent: 29 August 2019 22:08 > To: Roy Eastwood; samba at lists.samba.org > Subject: Re: [Samba] Online Backup Fails - list index out of range > > On 30/08/19 4:20 AM, Roy Eastwood via samba wrote: > > INFO 2019-08-29 17:04:06,889 pid:13945 /usr/lib/python3/dist-packages/samba/join.py #1643: Setting isSynchronized and > dsServiceName > > INFO 2019-08-29 17:04:07,303 pid:13945 /usr/lib/python3/dist-packages/samba/join.py #1555: Cloned domain MICROLYNX (SID > > S-1-5-21-4012640977-2272627666-3977488320) > > ERROR(<class 'IndexError'>): uncaught exception - list index out of range > > File "/usr/lib/python3/dist-packages/samba/netcmd/__init__.py", line 185, in _run > > return self.run(*args, **kwargs) > > File "/usr/lib/python3/dist-packages/samba/netcmd/domain_backup.py", line 258, in run > > new_sid = get_sid_for_restore(remote_sam) > > File "/usr/lib/python3/dist-packages/samba/netcmd/domain_backup.py", line 81, in get_sid_for_restore > > rid = int(res[0].get('rIDNextRID')[0]) > > > > Is this another bug? > > It's not really a bug. The backup process needs to allocate a new RID > and the DC you're running the backup against doesn't have a RID pool > allocated. E.g. see: > > https://lists.samba.org/archive/samba/2019-February/221619.html > > I have improved the error message you get in this case on master, but I > haven't backported this yet.Thanks for that. I still don't understand though as the DC on which I am doing the backup is the RIDAllocationMasterRole owner (actually it owns all the FSMO roles), so shouldn't it be able to provide a new RID? Or am I missing something? Roy
Andrew Bartlett
2019-Aug-30 00:30 UTC
[Samba] Online Backup Fails - list index out of range
On Thu, 2019-08-29 at 22:59 +0100, Roy Eastwood via samba wrote:> > -----Original Message----- > > From: Tim Beale [mailto:timbeale at catalyst.net.nz] > > Sent: 29 August 2019 22:08 > > To: Roy Eastwood; samba at lists.samba.org > > Subject: Re: [Samba] Online Backup Fails - list index out of range > > > > > > > Is this another bug? > > > > It's not really a bug. The backup process needs to allocate a new RID > > and the DC you're running the backup against doesn't have a RID pool > > allocated. E.g. see: > > > > https://lists.samba.org/archive/samba/2019-February/221619.html > > > > I have improved the error message you get in this case on master, but I > > haven't backported this yet. > > Thanks for that. I still don't understand though as the DC on which I am doing the backup is the RIDAllocationMasterRole owner (actually it owns all the FSMO roles), so shouldn't it be able to provide a new RID? Or am I missing something?It will do that when you create a user on that domain. We don't modify the domain we are backing up so we can't do this properly, instead we emulate the algorithm locally. The typical case is a backup of a DC that has just joined the domain, but recent Samba versions explicitly get a RID pool on join for reliability, so that shouldn't happen any more. Andrew Bartlett -- Andrew Bartlett https://samba.org/~abartlet/ Authentication Developer, Samba Team https://samba.org Samba Development and Support, Catalyst IT https://catalyst.net.nz/services/samba
> -----Original Message----- > From: Andrew Bartlett [mailto:abartlet at samba.org] > Sent: 30 August 2019 01:30 > To: Roy Eastwood; 'Tim Beale'; samba at lists.samba.org > Subject: Re: [Samba] Online Backup Fails - list index out of range > > On Thu, 2019-08-29 at 22:59 +0100, Roy Eastwood via samba wrote: > > > -----Original Message----- > > > From: Tim Beale [mailto:timbeale at catalyst.net.nz] > > > Sent: 29 August 2019 22:08 > > > To: Roy Eastwood; samba at lists.samba.org > > > Subject: Re: [Samba] Online Backup Fails - list index out of range > > > > > > > > > > Is this another bug? > > > > > > It's not really a bug. The backup process needs to allocate a new RID > > > and the DC you're running the backup against doesn't have a RID pool > > > allocated. E.g. see: > > > > > > https://lists.samba.org/archive/samba/2019-February/221619.html > > > > > > I have improved the error message you get in this case on master, but I > > > haven't backported this yet. > > > > Thanks for that. I still don't understand though as the DC on which I am doing the backup is the RIDAllocationMasterRole owner > (actually it owns all the FSMO roles), so shouldn't it be able to provide a new RID? Or am I missing something? > > It will do that when you create a user on that domain. We don't modify > the domain we are backing up so we can't do this properly, instead we > emulate the algorithm locally. > > The typical case is a backup of a DC that has just joined the domain, > but recent Samba versions explicitly get a RID pool on join for > reliability, so that shouldn't happen any more. > > Andrew BartlettThanks for the clarification. I added a new user to the domain, and repeated the backup and it completed successfully. Many thanks. Roy