Help! Lost Blocks from unpruned full chain after NFS-Disconnect

[BaggedEyes]

Hello,
this is my first post to this Forum, so please be gentle :-)

I run a full blockchain using an NFS-mounted share - after
a network-hiccup  blk00009.dat and  blk00067.dat are missing -
about 250 MByte... I dont want to download umteen Gigabytes -
is there a shorter way (reindex would pose no problem)?

Please help! Bag

[achow101]

A reindex may fix the problem. However you still may need to redownload the entire blockchain.

[BaggedEyes]

thanks for the prompt reply!
reindex fails at the
moment at the missing block. I have reason to
believe that > 700 blocks are fine and only 2 missing.
I want to avoid starting from scratch.
regards Bag

[jackg]

Hello,
this is my first post to this Forum, so please be gentle :-)

I run a full blockchain using an NFS-mounted share - after
a network-hiccup  blk00009.dat and  blk00067.dat are missing -
about 250 MByte... I dont want to download umteen Gigabytes -
is there a shorter way (reindex would pose no problem)?

Please help! Bag

I think this constitutes the definition of a corrupted blockchain.
You'll probably have to reindex unless you can find the files somewhere and the software doesn't notice (but that's unlikely - if you want them, I'll probably have those files as they are quite early in the blockchain)...

[achow101]

thanks for the prompt reply!
reindex fails at the
moment at the missing block. I have reason to
believe that > 700 blocks are fine and only 2 missing.
I want to avoid starting from scratch.
regards Bag

Well unfortunately right now it won't work if you are missing blocks, so you will need to redownload those blocks. I don't think there is any mechanism yet to redownload only the blocks that are missing, so you will need to redownload the whole thing.

[BaggedEyes]

Hello,
this is my first post to this Forum, so please be gentle :-)

I run a full blockchain using an NFS-mounted share - after
a network-hiccup  blk00009.dat and  blk00067.dat are missing -
about 250 MByte... I dont want to download umteen Gigabytes -
is there a shorter way (reindex would pose no problem)?

Please help! Bag

I think this constitutes the definition of a corrupted blockchain.
You'll probably have to reindex unless you can find the files somewhere and the software doesn't notice (but that's unlikely - if you want them, I'll probably have those files as they are quite early in the blockchain)...

Thank you so much jackg!

I think relplacing the files helped - currently reindexing and it did not abort (yet); I will
report success if "normal" operation commences.

Thanks again - if this works, it saved a lot of time...(and bandwidth).
Regards Bag

[jackg]

Hello,
this is my first post to this Forum, so please be gentle :-)

I run a full blockchain using an NFS-mounted share - after
a network-hiccup  blk00009.dat and  blk00067.dat are missing -
about 250 MByte... I dont want to download umteen Gigabytes -
is there a shorter way (reindex would pose no problem)?

Please help! Bag

I think this constitutes the definition of a corrupted blockchain.
You'll probably have to reindex unless you can find the files somewhere and the software doesn't notice (but that's unlikely - if you want them, I'll probably have those files as they are quite early in the blockchain)...

Thank you so much jackg!

I think relplacing the files helped - currently reindexing and it did not abort (yet); I will
report success if "normal" operation commences.

Thanks again - if this works, it saved a lot of time...(and bandwidth).
Regards Bag

That's great, hopefully it doesn't face the same problem. If you have already used it for a few hours then it should have no problems as it'll have accepted the 9th data file that was missing.

[calkob]

Hello,
this is my first post to this Forum, so please be gentle :-)

I run a full blockchain using an NFS-mounted share - after
a network-hiccup  blk00009.dat and  blk00067.dat are missing -
about 250 MByte... I dont want to download umteen Gigabytes -
is there a shorter way (reindex would pose no problem)?

Please help! Bag

I wouldn't stress to much about it man, the latest version of core is super fast compared to older versions.  Just get it started and forget about it check back in a week and i bet it will be done ..... 

[BaggedEyes]

I wouldn't stress to much about it man, the latest version of core is super fast compared to older versions.  Just get it started and forget about it check back in a week and i bet it will be done ..... 

Thank you all :-) - After replacing the two blocks the reindex completed in a few minutes - nevertheless it startet
to "rebuild" - I am not sure why. But anyway,  "calkob" is correct: it did not take as long as I expected.
For those interested,here is my monitoring of this complete reload (47 hours, if I calculated correctly :-)):

https://i.imgur.com/G69WgTW.png https://i.imgur.com/ykkujhw.png

http://imgur.com/a/CI4Jg