During R30 ( see https://elog.ph.ed.ac.uk/AIDA/154 ) obtain the following merger error messages
(rate ~ 10 per minute) with merger rate ~100-200k data items/s 

Merge: Timestamp (611b2080877)[Q 2, Stream 2] is older than previous (611c207f4ef) [Q 2, Stream 2]
Merge: Attempting resynchronisation to timestamp 611c207f4f0 or greater
Merge: Last sync pulse time: 611c20400a0
Merge: Last timestamp seen: 611c207f4ef
Merge: Restarting at timestamp 611c72800a0
Merge: Timestamp (611f6f0101f)[Q 2, Stream 2] is older than previous (61206effe27) [Q 2, Stream 2]
Merge: Attempting resynchronisation to timestamp 61206effe28 or greater
Merge: Last sync pulse time: 61206ec00a0
Merge: Last timestamp seen: 61206effe27
Merge: Restarting at timestamp 612088000a0
Merge: Timestamp (61213ac1e1f)[Q 2, Stream 2] is older than previous (61223abffa7) [Q 2, Stream 2]
Merge: Attempting resynchronisation to timestamp 61223abffa8 or greater
Merge: Last sync pulse time: 61223a800a0
Merge: Last timestamp seen: 61223abffa7
Merge: Restarting at timestamp 6122ab400a0

611c72800a0-611c207f4ef = 5200BB1 = 85986225 * 1e-8 = 0.86s

612088000a0-61206effe27 = 1900279 = 0.26s 

6122ab400a0-61223abffa7 = 70800F9 = 1.2s

Resynchronisation appears to require significant periods of time (~s) - why?

I assume this implies significant dead time (data loss) too?