p2p: Hardcoded seeds update pre-0.15 branch #10948

1. 
   laanwj commented at 10:43 AM on July 28, 2017: member

   Just the usual

       curl -s http://bitcoin.sipa.be/seeds.txt.gz | gzip -dc > seeds_main.txt
       python3 makeseeds.py < seeds_main.txt > nodes_main.txt
       python3 generate-seeds.py . > ../../src/chainparamsseeds.h
   

2. laanwj added the label P2P on Jul 28, 2017
3. laanwj added this to the milestone 0.15.0 on Jul 28, 2017
4. 
   fanquake commented at 10:49 AM on July 28, 2017: member

   utACK f5b0980

5. 
   achow101 commented at 8:33 PM on July 28, 2017: member

   utACK f5b09807e5eb21a72602a4a24a3892ebcf1ae244

6. 
   gmaxwell commented at 12:05 AM on July 29, 2017: contributor

   Needs an update to include 0.14 nodes.

   -PATTERN_AGENT = re.compile(r"^(/Satoshi:0.12.(0|1|99)/|/Satoshi:0.13.(0|1|2|99)/)$") +PATTERN_AGENT = re.compile(r"^(/Satoshi:0.13.(1|2|99)/|/Satoshi:0.14.(0|1|2|99)/)$")

   My proposal leaves out 0.13.0 due to no segwit support; they're under 8% of the 0.13.1+ nodes.

7. 
   laanwj commented at 11:06 AM on July 30, 2017: member

   After the version bump, master will be 0.15.99 - shouldn't we include 0.15.x as well? Otherwise we're likely to forget it if we do another seeds update after the version bump for some reason.

8. 
   MarcoFalke commented at 11:16 AM on July 30, 2017: member

   According to the release process, the seed update is only done before every major release. So it likely doesn't matter.

9. 
   laanwj commented at 11:19 AM on July 30, 2017: member

   Ok... (updated)

10. p2p: Hardcoded seeds update pre-0.15 branch f0aceddf88
11. laanwj force-pushed on Jul 30, 2017
12. 
    gmaxwell commented at 7:58 AM on July 31, 2017: contributor

    @laanwj I would have been fine sticking 0.15 stuff in there, but I think it doesn't matter.

13. gmaxwell approved
14. 
    gmaxwell commented at 7:58 AM on July 31, 2017: contributor

    utACK

15. 
    MarcoFalke commented at 8:16 AM on July 31, 2017: member

    ACK f0acedd. (I ran the same commands and a subset of nodes was identical)

16. 
    jonasschnelli commented at 8:59 AM on July 31, 2017: contributor

    Tested the ipv4 addresses (as indication if the source data is up to date): Connect (timeout 6s), handshake and get headers with locator at height 161280. 420 of the 512 passed this test (~82%).

    partial tested ACK: f0aceddf88d7d680d3b55d302b8b2935f4b1c53f

17. 
    laanwj commented at 9:32 AM on July 31, 2017: member

    Thanks for testing @MarcoFalke @jonasschnelli !

    I would have been fine sticking 0.15 stuff in there, but I think it doesn't matter.

    Yes, I agree it doesn't matter, I'm a little bit scared that the version filtering regex will bite us some day by accident, thanks for reminding me of it anyhow.

    Edit: maybe a good idea (for the future, not for this PR) would be to make it a required command-line option. This forces thinking about it when issuing the command.

18. laanwj merged this on Jul 31, 2017
19. laanwj closed this on Jul 31, 2017

    ---

20. laanwj referenced this in commit df3a6f4ee4 on Jul 31, 2017
21. MarcoFalke locked this on Sep 8, 2021

Milestone
0.15.0