On Thu, 5 Oct 2023 at 18:48, Justin Pryzby <pryzby@telsasoft.com> wrote:
>
> On an instance running pg16.0:
>
> log_time | 2023-10-05 10:03:00.014-05
> backend_type | autovacuum worker
> left | page verification failed, calculated checksum 5074 but expected 5050
> context | while scanning block 119 of relation "public.postgres_log_2023_10_05_0900"
>
> This is the only error I've seen so far, and for all I know there's a
> issue on the storage behind the VM, or a cosmic ray hit. But I moved
> the table out of the way and saved a copy of get_raw_page() in case
> someone wants to ask about it.
>
> postgres=# SELECT * FROM heap_page_item_attrs(get_raw_page(801594131::regclass::text, 119), 801594131);
> lp | lp_off | lp_flags | lp_len | t_xmin | t_xmax | t_field3 | t_ctid | t_infomask2 | t_infomask | t_hoff | t_bits
|t_oid | t_attrs
> 1 | 2304 | 1 | 16 | | | | | | | |
| |
> 2 | 8160 | 1 | 16 | | | | | | | |
| |
> 3 | 8144 | 1 | 16 | | | | | | | |
| |
> ...all the same except for lp_off...
> 365 | 2352 | 1 | 16 | | | | | | | |
| |
> 366 | 2336 | 1 | 16 | | | | | | | |
| |
> 367 | 2320 | 1 | 16 | | | | | | | |
| |
That's not a HEAP page; it looks more like a btree page: lp_len is too
short for heap (which starts at lp_len = 24), and there are too many
line pointers for an 8KiB heap page. btree often has lp_len of 16: 8
bytes indextuple header, one maxalign of data (e.g. int or bigint).
So, assuming it's a block of a different relation kind, then it's also
likely it was originally located elsewhere in that other relation,
indeed causing the checksum failure. You can further validate this by
looking at the page header's pd_special value - if it is 8176, that'd
be another indicator for it being a btree.
Kind regards,
Matthias van de Meent.
