One of the issues with ausearch's checkpoint code is how to recover from failures. A classic failure is to perform a checkpoint on a busy system and then delay too long before running the next invocation of ausearch and as a result of the delay, the checkpointed event cannot be found in the files in /var/log/audit. There are other failures, such as re-use of inodes etc. For those of you who haven't noted the ausearch --checkpoint change, it basically records the details of the last complete audit event it processed or printed in a checkpoint file. It records not only the event time, but also the event node, serial, type and the file device and inode. Thus, when you next invoke ausearch with this option, the next event to process is the next complete event since the one recorded. Should an error occur when attempting to find the next complete event to process, ausearch will exit. At this point, I believe the best recovery action is to extract only the event time from the checkpoint file and ask for all complete events after that time (i.e. as opposed to the usual action of comparing time, event id, type, log file details etc).

There are at last two solutions: a. We can patch ausearch to take a --checkpoint-time-only flag which means ausearch will look for all events since the time in the checkpoint file. This provides the best granularity in time as it goes down to msecs.

b. We extract the timestamp from the checkpoint file, convert it to a date and time and use ausearch's --start option to find all events since the time in the checkpoint file. The first provides greater granularity in time as it goes to msecs.

I can provide a patch. Do you want it?

Just to confirm: the patch would modify the --start command line processing to accept a string argument of 'checkpoint-time' AND if a checkpoint file has also been provided via the --checkpoint arg AND there is a timestamp within the specified file, we use the timestamp stored within the file?

On Mon, 2014-08-18 at 14:13 -0400, Steve Grubb wrote: > Hello, > > On Saturday, August 16, 2014 09:25:16 AM Burn Alting wrote: > > One of the issues with ausearch's checkpoint code is how to recover from > > failures. A classic failure is to perform a checkpoint on a busy system > > and then delay too long before running the next invocation of ausearch > > and as a result of the delay, the checkpointed event cannot be found in > > the files in /var/log/audit. There are other failures, such as re-use of > > inodes etc. > > > > For those of you who haven't noted the ausearch --checkpoint change, it > > basically records the details of the last complete audit event it > > processed or printed in a checkpoint file. It records not only the event > > time, but also the event node, serial, type and the file device and > > inode. Thus, when you next invoke ausearch with this option, the next > > event to process is the next complete event since the one recorded. > > > > Should an error occur when attempting to find the next complete event to > > process, ausearch will exit. At this point, I believe the best recovery > > action is to extract only the event time from the checkpoint file and > > ask for all complete events after that time (i.e. as opposed to the > > usual action of comparing time, event id, type, log file details etc). > > Would anyone be opposed to making that the default behavior? > > > There are at last two solutions: > > a. We can patch ausearch to take a --checkpoint-time-only flag which > > means ausearch will look for all events since the time in the checkpoint > > file. This provides the best granularity in time as it goes down to > > msecs. > > I am worried about the proliferation of command line switches. I'd rather > make a new --start target. e.g. --start checkpoint-time. > > > b. We extract the timestamp from the checkpoint file, convert it to a > > date and time and use ausearch's --start option to find all events since > > the time in the checkpoint file. > > > > The first provides greater granularity in time as it goes to msecs. > > If one is the timestamp of the file, that might be misleading. I don't > know if touching a file is an auditable event. No time to investigate > right now either. I'd rather see the time taken from within the file. > > > I can provide a patch. Do you want it? > > Sure, if its based on a --start target. > > -Steve

This makes sense to me. I am all for it. +1 R, -Joe ______________________________________________________________ From: Steve Grubb <sgrubb(a)redhat.com&gt; To: burn(a)swtf.dyndns.org Cc: linux-audit(a)redhat.com Sent: Monday, August 18, 2014 5:59 PM Subject: Re: ausearch checkpoint capability Hello, On Tuesday, August 19, 2014 07:49:50 AM Burn Alting wrote: > Just to confirm: > > the patch would modify the --start command line processing to accept > a string argument of 'checkpoint-time' AND if a checkpoint file has also > been provided via the --checkpoint arg AND there is a timestamp within > the specified file, we use the timestamp stored within the file? Yes. I am close to doing a new release of the audit package. I am kind of aiming towards the end of this week. If its ready by then, I'll include it in the new release. If not, maybe next release. Also, if anyone else has bugs to report, patches to send, etc. now would be a good time if they needed it to go out soonish. Thanks, -Steve > On Mon, 2014-08-18 at 14:13 -0400, Steve Grubb wrote: > > Hello, > > > > On Saturday, August 16, 2014 09:25:16 AM Burn Alting wrote: > > > One of the issues with ausearch's checkpoint code is how to recover from > > > failures. A classic failure is to perform a checkpoint on a busy system > > > and then delay too long before running the next invocation of ausearch > > > and as a result of the delay, the checkpointed event cannot be found in > > > the files in /var/log/audit. There are other failures, such as re-use of > > > inodes etc. > > > > > > For those of you who haven't noted the ausearch --checkpoint change, it > > > basically records the details of the last complete audit event it > > > processed or printed in a checkpoint file. It records not only the event > > > time, but also the event node, serial, type and the file device and > > > inode. Thus, when you next invoke ausearch with this option, the next > > > event to process is the next complete event since the one recorded. > > > > > > Should an error occur when attempting to find the next complete event to > > > process, ausearch will exit. At this point, I believe the best recovery > > > action is to extract only the event time from the checkpoint file and > > > ask for all complete events after that time (i.e. as opposed to the > > > usual action of comparing time, event id, type, log file details etc). > > > > Would anyone be opposed to making that the default behavior? > > > > > There are at last two solutions: > > > a. We can patch ausearch to take a --checkpoint-time-only flag which > > > means ausearch will look for all events since the time in the checkpoint > > > file. This provides the best granularity in time as it goes down to > > > msecs. > > > > I am worried about the proliferation of command line switches. I'd rather > > make a new --start target. e.g. --start checkpoint-time. > > > > > b. We extract the timestamp from the checkpoint file, convert it to a > > > date and time and use ausearch's --start option to find all events since > > > the time in the checkpoint file. > > > > > > The first provides greater granularity in time as it goes to msecs. > > > > If one is the timestamp of the file, that might be misleading. I don't > > know if touching a file is an auditable event. No time to investigate > > right now either. I'd rather see the time taken from within the file. > > > > > I can provide a patch. Do you want it? > > > > Sure, if its based on a --start target. > > > > -Steve -- Linux-audit mailing list Linux-audit(a)redhat.com https://www.redhat.com/mailman/listinfo/linux-audit