can it run with --db.pagesize=8kb ?

[flywukong]

the mdbx should be consistent with the config of pagesize , so this tarball is done with --db.pagesize=8kb or --db.pagesize=4kb ? if it is runned with 4kb , I am afraid that the data can not be run with flag --db.pagesize=8kb

PS : I did not use your script to run it , I have run node with below command and it works well for 10+ hours, ./erigon --chain=bsc --datadir /server/archive --snap.keepblocks --db.pagesize=8kb

[allada]

Currently it is 4k (default), but I've been working on a new sync that is 16k that should be released in a week or so.

[flywukong]

we can also use mdbx_dump/mdbx_load tools to convert the data into 8k or 16k

[allada]

As of https://github.com/allada/bsc-archive-snapshot/commit/4d8d56ac1fe424c119ef1925f387ce24df484e31, it uses 16k blocks. This solved much of the "can't keep up with latest blocks issues".

[allada]

Currently it is 4k (default), but I've been working on a new sync that is 16k that should be released in a week or so.

On Thu, Jul 21, 2022, 9:25 PM flywukong @.***> wrote:

the mdbx should be consistent with the config of pagesize , so this tarball is done with --db.pagesize=8kb or --db.pagesize=4kb ?

— Reply to this email directly, view it on GitHub https://github.com/allada/bsc-archive-snapshot/issues/25, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAN7CIT44BGIK6XFIIBZ2ATVVIBCJANCNFSM54JUGKRA . You are receiving this because you are subscribed to this thread.Message ID: @.***>