Open vncoelho opened 6 years ago
@vncoelho The block time is given by this sql query:
https://github.com/CityOfZion/neo-scan/blob/master/apps/neoscan/lib/neoscan/blocks/blocks.ex#L92
would it be possible for you to check in the postgres database to check the block time and see if the time is correct (block with higher index have always a higher timestamp)? Would it be possible that different neo nodes have different clocks and are not registering the same time for a block?
I will, @adrienmo, what is the best way to query this once I am inside the postgres docker?
But the timestamp on the screenshot looks correct, that is the strange thing.
psql -h 127.0.0.1 -U postgres -d neoscan_prodv password: postgres
And then you can run this command:
select index,time,extract(epoch FROM (time - lead(time) OVER ()))::integer as lag from blocks order by index desc limit 15;
It's indeed weird that on the screenshot everything looks correct for the time.
I am trying to replicate, as soon as it happens I will update here, @adrienmo.
Thanks for the tip and attention.
@adrienmo, How to select a specific epoch in that command you sent above?
73 | 2018-12-13 12:42:30 | 1544704950
72 | 2018-12-13 12:42:27 | 1544704947
71 | 2018-12-13 12:42:24 | 1544704944
70 | 2018-12-13 12:42:20 | 1544704940
69 | 2018-12-13 12:42:17 | 1544704937
68 | 2018-12-13 12:42:14 | 1544704934
67 | 2018-12-13 12:42:11 | 1544704931
66 | 2018-12-13 12:42:08 | 1544704928
65 | 2018-12-13 12:42:05 | 1544704925
64 | 2018-12-13 12:42:02 | 1544704922
63 | 2018-12-13 12:41:59 | 1544704919
62 | 2018-12-13 12:41:55 | 1544704915
61 | 2018-12-13 12:41:52 | 1544704912
60 | 2018-12-13 12:41:48 | 1544704908
59 | 2018-12-13 12:41:45 | 1544704905
58 | 2018-12-13 12:41:42 | 1544704902
57 | 2018-12-13 12:41:39 | 1544704899
56 | 2018-12-13 12:41:36 | 1544704896
55 | 2018-12-13 12:41:33 | 1544704893
54 | 2018-12-13 12:41:30 | 1544704890
53 | 2018-12-13 12:41:27 | 1544704887
52 | 2018-12-13 12:41:24 | 1544704884
Done, I just increased the number being printed.
There should be something in the calculus, it just happens in some pages of our privatenet.