I agree with Ted, these are both definitely bugs. On Thu, Oct 20, 2011 at 6:52 AM, Ted Dunning <[email protected]> wrote:
> Both of these sound like bugs. > > Sent from my iPhone > > On Oct 20, 2011, at 6:44, Thomas Koch <[email protected]> wrote: > > > Hi, > > > > I'm woundering: The check operation is threaded like a write operation in > > PrepRequestProcessor. It gets a TxnHeader attached and increments the > zxid. > > However the switch cases in ObserverRequestProcessor, CommitProcessor, > > FollowerRequestProcessor and ReadOnlyRequestProcessor thread a check like > a > > read operation. > > I don't know, whether this is a bug, but the inconsistency makes one > > suspicious and merits an explaining comment. > > > > Btw. The switch block in the ReadOnlyRequestProcessor does not list multi > as a > > write operation. Can this cause problems? > > > > Best regards, > > > > Thomas Koch, http://www.koch.ro >
