Re: pg_resetwal: Corrections around -c option

2023-10-10 Thread Alvaro Herrera
On 2023-Oct-10, Peter Eisentraut wrote: > On 09.10.23 17:48, Alvaro Herrera wrote: > > Hmm, not sure about this. [...] > > Would those issues also apply to the other SLRU-based guides on this man > page? Are they all a bit wrong? I didn't verify, but I think it's likely that they do and they a

Re: pg_resetwal: Corrections around -c option

2023-10-10 Thread Peter Eisentraut
On 09.10.23 17:48, Alvaro Herrera wrote: Hmm, not sure about this. SLRU_PAGES_PER_SEGMENT is 32, and COMMIT_TS_XACTS_PER_PAGE is 819, so this formula gives decimal 26208 = 0x6660. But more generally, I'm not sure about the algorithm. Really, the safe value also depends on how large the latest

Re: pg_resetwal: Corrections around -c option

2023-10-09 Thread Alvaro Herrera
On 2023-Sep-28, Peter Eisentraut wrote: > Branching off from [0], here is a for-discussion patch about some > corrections for the pg_resetwal -c (--commit-timestamp-ids) option. > > First, in the documentation for finding a manual value for the -c option > based on the files present in the data d