Is there a "strategic" plan for the tool?
Will it continue to grow in supporting disk functionality?
Will it all be in one executable?
I like your "mission statement" - "Simplicity is prerequisite for reliability", but being a user from the old v1 I have noticed a growth in functionality.
Have you considered to use separate executables for the core hashing & parity file-protection function from the supporting functionalities, both using the content file (pool, dup and such) and unrelated (smart)?
With smart-functions i would assume agility is of high importance, but with core-parity robustness and reliability is of most essence. The separation would also allow more people be able to provide support with different knowledge areas and experience as it is clear separation without having to know all.
When talking about core-function (with risk of contradicting my self):
From the past I know there have been requests of using snapraid's great hashing function without parity needs and also allow hashing validation on an off-site data replica without the parity disk.
Is this function not supported because of "strategic" reasons or from a not-have-time-to-code-it reason? (I made a quick-n-dirty mod on 7.1 with very small number of changes as you already have an easy well structured source code, but is missing testing and all other nice things you have. It actually supported both requests and is much nicer than using "rsyncsums".)
Regards
/X
Added source for modding 7.1 as "base" if anyone else would be interrested to work on it. Do not use unless you understand it completely.
Hi, (thanks for a very useful tool)
Is there a "strategic" plan for the tool?
Will it continue to grow in supporting disk functionality?
Will it all be in one executable?
I like your "mission statement" - "Simplicity is prerequisite for reliability", but being a user from the old v1 I have noticed a growth in functionality.
Have you considered to use separate executables for the core hashing & parity file-protection function from the supporting functionalities, both using the content file (pool, dup and such) and unrelated (smart)?
With smart-functions i would assume agility is of high importance, but with core-parity robustness and reliability is of most essence. The separation would also allow more people be able to provide support with different knowledge areas and experience as it is clear separation without having to know all.
When talking about core-function (with risk of contradicting my self):
From the past I know there have been requests of using snapraid's great hashing function without parity needs and also allow hashing validation on an off-site data replica without the parity disk.
Is this function not supported because of "strategic" reasons or from a not-have-time-to-code-it reason? (I made a quick-n-dirty mod on 7.1 with very small number of changes as you already have an easy well structured source code, but is missing testing and all other nice things you have. It actually supported both requests and is much nicer than using "rsyncsums".)
Regards
/X
Added source for modding 7.1 as "base" if anyone else would be interrested to work on it. Do not use unless you understand it completely.
Last edit: xad 2015-04-29