[URGENT] Delegates Please Revert Upgrade!

https://bitsharestalk.org/index.php?topic=7067.msg161312#msg161312

https://bitsharestalk.org/index.php?topic=7067.msg161432#msg161432

All community members please help spread the word.

Quote from: bytemaster on December 12, 2014, 03:48:39 PM
All delegates should downgrade to the last (Hot Fix) 0.4.24.1 immediately to rejoin the majority of users and start confirming that network.   

I *knew* I should have waited for official word.

So, downgrade it is...

Confused ...

Quote from: bytemaster on December 12, 2014, 03:48:39 PM
All users who haven't upgraded to the pre-release are on the minority fork. 

All delegates should downgrade to the last (Hot Fix) 0.4.24.1 immediately to rejoin the majority of users and start confirming that network.   

0.4.25-RC2 had an intractable issue.

Vikram knows the details.   We are working on a new upgrade but want to get the network back to a stable condition for users as soon as possible.   

All valid transactions have been applied to both networks.

You don't mean the delegates that run v0.4.25 I suppose...

All users who haven't upgraded to the pre-release are on the minority fork. 

All delegates should downgrade to the last (Hot Fix) 0.4.24.1 immediately to rejoin the majority of users and start confirming that network.   

0.4.25-RC2 had an intractable issue.

Vikram knows the details.   We are working on a new upgrade but want to get the network back to a stable condition for users as soon as possible.   

All valid transactions have been applied to both networks.

Edit by Vikram: Please read https://bitsharestalk.org/index.php?topic=12207.msg161341#msg161341

Quote from: biophil on December 12, 2014, 03:34:10 PM
I'm thinking maybe I'm on a fork; "get_info" says that delegate participation is 4.43% and I'm stuck downloading blocks at 1246468. Will I have to re-download the blockchain? If so, how do I ensure that I ultimately end up on the right fork?

Or maybe the answers to this question are too complicated and I should just wait for smarter people to fix everything...

run with --rebuild-index first, before you try --resync-blockchain as a last resort.

Quote from: emski on December 12, 2014, 03:31:58 PM

Quote from: Riverhead on December 12, 2014, 03:23:30 PM

Quote from: svk on December 12, 2014, 03:04:14 PM

Quote from: xeroc on December 12, 2014, 02:57:49 PM
my delegates are finally producing again .. sorry for the delay .. was sleeping and have noticed the EARLY!!! hardfork!
also had to switch to a different machine .. I hope the devs can fix the memory leak so that I can swtich back to the original machine ..

Fortunately thanks to all the missed blocks the time of the hardfork has been pushed back by over two hours! ;)

Oh and there's no need to change machines finally, I did so at first but turns out you can reindex and resync just fine with 512mb RAM and a swap. It's the compile part that needs lots of RAM.

Weird. My experience was more like xeroc's. The machine locked up so bad the live delegate was missing blocks.

I saw some unusual crashes at client start taking much CPU just before the crash. Other than this the update went smoothly.

Ya, seems to have gone better for some than others for sure. Network was down to 60% for a bit.

I'm thinking maybe I'm on a fork; "get_info" says that delegate participation is 4.43% and I'm stuck downloading blocks at 1246468. Will I have to re-download the blockchain? If so, how do I ensure that I ultimately end up on the right fork?

Or maybe the answers to this question are too complicated and I should just wait for smarter people to fix everything...

Quote from: Riverhead on December 12, 2014, 03:23:30 PM

Quote from: svk on December 12, 2014, 03:04:14 PM

Quote from: xeroc on December 12, 2014, 02:57:49 PM
my delegates are finally producing again .. sorry for the delay .. was sleeping and have noticed the EARLY!!! hardfork!
also had to switch to a different machine .. I hope the devs can fix the memory leak so that I can swtich back to the original machine ..

Fortunately thanks to all the missed blocks the time of the hardfork has been pushed back by over two hours! ;)

Oh and there's no need to change machines finally, I did so at first but turns out you can reindex and resync just fine with 512mb RAM and a swap. It's the compile part that needs lots of RAM.

Weird. My experience was more like xeroc's. The machine locked up so bad the live delegate was missing blocks.

I saw some unusual crashes at client start taking much CPU just before the crash. Other than this the update went smoothly.

Quote from: svk on December 12, 2014, 03:04:14 PM
Fortunately thanks to all the missed blocks the time of the hardfork has been pushed back by over two hours! ;)

How is that fortunate? We already suffered the most brutal part of hard forks unintentionally, at this point we should want block 1249400 to arrive as soon as possible.

Quote from: svk on December 12, 2014, 03:04:14 PM

Quote from: xeroc on December 12, 2014, 02:57:49 PM
my delegates are finally producing again .. sorry for the delay .. was sleeping and have noticed the EARLY!!! hardfork!
also had to switch to a different machine .. I hope the devs can fix the memory leak so that I can swtich back to the original machine ..

Fortunately thanks to all the missed blocks the time of the hardfork has been pushed back by over two hours! ;)

Oh and there's no need to change machines finally, I did so at first but turns out you can reindex and resync just fine with 512mb RAM and a swap. It's the compile part that needs lots of RAM.

Weird. My experience was more like xeroc's. The machine locked up so bad the live delegate was missing blocks.

I think develop should lunch new version in morning in case fatal bug

there are so many translation now , is it normal

It will be nice when we dont hard fork as often, and when everything is run for a while first on devshares to prevent this. :)

Quote from: svk on December 12, 2014, 03:04:14 PM

Quote from: xeroc on December 12, 2014, 02:57:49 PM
my delegates are finally producing again .. sorry for the delay .. was sleeping and have noticed the EARLY!!! hardfork!
also had to switch to a different machine .. I hope the devs can fix the memory leak so that I can swtich back to the original machine ..

Fortunately thanks to all the missed blocks the time of the hardfork has been pushed back by over two hours! ;)

Oh and there's no need to change machines finally, I did so at first but turns out you can reindex and resync just fine with 512mb RAM and a swap. It's the compile part that needs lots of RAM.

gonna give it a shot ..
redownloading blockchain on it ..
thanks for the hint ..

Quote from: monsterer on December 12, 2014, 03:01:59 PM

Quote from: xeroc on December 12, 2014, 02:57:49 PM
my delegates are finally producing again .. sorry for the delay .. was sleeping and have noticed the EARLY!!! hardfork!
also had to switch to a different machine .. I hope the devs can fix the memory leak so that I can swtich back to the original machine ..

What a nice surprise you woke up to! :)

So, is the memory leak on-going, or just at re-indexing time?

the client reindexes just fine .. then comes up completely and freezes with 99%CPU and over 2GB ram consumption .. not matter how often I restart the client