Careful when upgrading delegate to 0.4.25

[vikram]

[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.

[abit]

It seems that a fork occurred before expected (<1249400).
All delegates should upgrade. Other users should upgrade as well. Wish the binary releases asap.
@vikram.

[svk]

I've updated the release notes to include a note about the memory requirement: https://github.com/BitShares/bitshares/releases/tag/v0.4.25

This doesn't appear to be true for reindexing and syncing, I've successfully reindexed and sync my seednode which has 512mb RAM and 4gb swap.

I compiled it on my personal computer which has 8gb of RAM so no issues there.

[ripplexiaoshan]

I am using 2GB VPS, and it seems fine for both compiling and running, but my delegate keeps missing blocks while the wallet is up，according to bitsharesblocks.com.

[emski]

0.4.24 instances report a lot of duplicate transactions when blockchain_use_forks is used. Is it the same with you ?

[svk]

Our delegate is running fine, but it is consuming a lot of virtual memory (2.3GB, but doesn't seem to be increasing) during initial sync. My best guess is that the people with problems are swapping due to insufficient memory. We already suspected a leak in the previous client during syncing (last client required 1.6GB on windows for initial sync).

Are the forking issues related to the new fork resolution code? Maybe the fact that the old and the new are live at the same time?

[abit]

EDIT.
Compiling.

[btswildpig]

I've updated the release notes to include a note about the memory requirement: https://github.com/BitShares/bitshares/releases/tag/v0.4.25

will delete the blockchain before upgrade helps ?

[vikram]

I've updated the release notes to include a note about the memory requirement: https://github.com/BitShares/bitshares/releases/tag/v0.4.25

[Riverhead]

yep, I think 3 to 4GB is a good minimum requirement for now for delegates, until we can isolate the cause of the heavier memory requirements.

Agreed. Going to upgrade.

[dannotestein]

yep, I think 3 to 4GB is a good minimum requirement for now for delegates, until we can isolate the cause of the heavier memory requirements.

[Xeldal]

My 1GB instance is continually crashing while syncing, Initially this crashed and (appeared to be) locked up.  I had to restart

My 4GB instance hasn't crashed at all.

[Riverhead]

It's possible the box may have come back. The delegate I had running on that box started missing blocks because nothing else could be processed.

I have it syncing on another box now and once it's all up to date I'll turn that on as the block signer and update my version number.

[dannotestein]

Our delegate is running fine, but it is consuming a lot of virtual memory (2.3GB, but doesn't seem to be increasing) during initial sync. My best guess is that the people with problems are swapping due to insufficient memory. We already suspected a leak in the previous client during syncing (last client required 1.6GB on windows for initial sync).

[Riverhead]

PM me your skype id?

[vikram]

Sorry for the short notice with the update but we had some important fixes.

Were you guys able to get these up and running at all?

In the past I've had my console freeze up during initial syncing, but comes back after syncing.

Was is crashes or just freezing? Any more details?

I am on Skype right now also.

[Riverhead]

client frozen on VPS with 2GB ram and 4GB swap .. :(
rand for a few seconds .. than stuck .. no more output .. not taking commands

Same. I had to shut the machine down via the web console of the VPS because the ssh session had completely froze.

[xeroc]

client frozen on VPS with 2GB ram and 4GB swap .. :(
rand for a few seconds .. than stuck .. no more output .. not taking commands

Code Select Expand


  PID USER      PR  NI    VIRT    RES    SHR S %CPU %MEM     TIME+ COMMAND                                     
23578 delegate  20   0 1664208 267832  25384 R 96.8 13.1   3:27.85 bitshares_client


Code Select Expand


             total       used       free     shared    buffers     cached
Mem:       2044004    2027600      16404          8       2456    1725848
-/+ buffers/cache:     299296    1744708
Swap:      4194300     408448    3785852


gonna stick with the backup machine for now ..

paging @vikram

[Riverhead]

Any help if run in limited-resource mode? I have not finish compiling.

I'll give that a try. I've gone back to 0.2.24 until I do more testing on dev server. I need to have my head examined for updating prod untested :(.

[abit]

Any help if run in limited-resource mode? I have not finish compiling.

[Riverhead]

I experienced something similuar ... it's a 2GB machine with 4GB swap space .. still the console froze to death with the RC2 build ... building stable atm ..

This was running the client. It core dumped once and then next run it started to download and then froze the machine. I've rebooted it from the console web page but sadly my other machines are reindexing and I can't bring up the delegate. Learning process :).

[xeroc]

I experienced something similuar ... it's a 2GB machine with 4GB swap space .. still the console froze to death with the RC2 build ... building stable atm ..

[Riverhead]

Perhaps the new high speed blockchain downloading or something but it hosed my VPS and crashed the 0.4.24 version (machine is currently locked up :P )