zebra.lucky
/
joinmarket-clientserver
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Browse Source

Don't use honest_makers in case: mempool conflict 

Prior to this commit, if a tumbler coinjoin negotiation
failed in Phase 2, then the retry as per the logic in
taker_utils.tumbler_taker_finished_update would always
attempt to retry the transaction with those counterparties
that returned valid !sig responses. However this ignored
the case that all the counterparties responded validly,
but there was a mempool conflict in the created transaction.
After this commit, if it is detected that all counterparties
responded, it is assumed that a mempool conflict or similar
occurred with the transaction, and therefore it is better
to fallback to a schedule tweak and choose randomly again,
not to fix the counterparty set (which is likely to result
in failing again).
master
Adam Gibson 6 years ago
parent
f174f7785e
commit
4af9c422a4
No known key found for this signature in database
GPG Key ID: 141001A1AF77F20B
1 changed files with 43 additions and 34 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 77
      jmclient/jmclient/taker_utils.py

77
jmclient/jmclient/taker_utils.py
Unescape View File

@ -250,46 +250,55 @@ def tumbler_taker_finished_update(taker, schedulefile, tumble_log, options,
tumble_log.info(waiting_message) tumble_log.info(waiting_message)
log.info(waiting_message) log.info(waiting_message)
else: else:
#a transaction failed, either because insufficient makers # a transaction failed, either because insufficient makers
#(acording to minimum_makers) responded in Phase 1, or not all # (acording to minimum_makers) responded in Phase 1, or not all
#makers responded in Phase 2. We'll first try to repeat without the # makers responded in Phase 2, or the tx was a mempool conflict.
#troublemakers. # If the tx was a mempool conflict, we should restart with random
# maker choice as usual. If someone didn't respond, we'll try to
# repeat without the troublemakers.
log.info("Schedule entry: " + str( log.info("Schedule entry: " + str(
taker.schedule[taker.schedule_index]) + \ taker.schedule[taker.schedule_index]) + \
" failed after timeout, trying again") " failed after timeout, trying again")
taker.add_ignored_makers(taker.nonrespondants) taker.add_ignored_makers(taker.nonrespondants)
#Is the failure in Phase 2? #Is the failure in Phase 2?
if not taker.latest_tx is None: if not taker.latest_tx is None:
#Now we have to set the specific group we want to use, and hopefully if len(taker.nonrespondants) == 0:
#they will respond again as they showed honesty last time. # transaction was created validly but conflicted in the
#Note that we must wipe the list first; other honest makers needn't # mempool; just try again without honest settings;
#have the right settings (e.g. max cjamount), so can't be carried # i.e. fallback to same as Phase 1 failure.
#over from earlier transactions. log.info("Invalid transaction; possible mempool conflict.")
taker.honest_makers = [] else:
taker.add_honest_makers(list(set( #Now we have to set the specific group we want to use, and hopefully
taker.maker_utxo_data.keys()).symmetric_difference( #they will respond again as they showed honesty last time.
set(taker.nonrespondants)))) #Note that we must wipe the list first; other honest makers needn't
#If insufficient makers were honest, we can only tweak the schedule. #have the right settings (e.g. max cjamount), so can't be carried
#If enough were, we prefer the restart with them only: #over from earlier transactions.
log.info("Inside a Phase 2 failure; number of honest respondants was: " + str(len(taker.honest_makers))) taker.honest_makers = []
log.info("They were: " + str(taker.honest_makers)) taker.add_honest_makers(list(set(
if len(taker.honest_makers) >= jm_single().config.getint( taker.maker_utxo_data.keys()).symmetric_difference(
"POLICY", "minimum_makers"): set(taker.nonrespondants))))
tumble_log.info("Transaction attempt failed, attempting to " #If insufficient makers were honest, we can only tweak the schedule.
"restart with subset.") #If enough were, we prefer to restart with them only:
tumble_log.info("The paramaters of the failed attempt: ") log.info("Inside a Phase 2 failure; number of honest "
tumble_log.info(str(taker.schedule[taker.schedule_index])) "respondants was: " + str(len(taker.honest_makers)))
#we must reset the number of counterparties, as well as fix who they log.info("They were: " + str(taker.honest_makers))
#are; this is because the number is used to e.g. calculate fees. if len(taker.honest_makers) >= jm_single().config.getint(
#cleanest way is to reset the number in the schedule before restart. "POLICY", "minimum_makers"):
taker.schedule[taker.schedule_index][2] = len(taker.honest_makers) tumble_log.info("Transaction attempt failed, attempting to "
retry_str = "Retrying with: " + str(taker.schedule[ "restart with subset.")
taker.schedule_index][2]) + " counterparties." tumble_log.info("The paramaters of the failed attempt: ")
tumble_log.info(retry_str) tumble_log.info(str(taker.schedule[taker.schedule_index]))
log.info(retry_str) #we must reset the number of counterparties, as well as fix who they
taker.set_honest_only(True) #are; this is because the number is used to e.g. calculate fees.
taker.schedule_index -= 1 #cleanest way is to reset the number in the schedule before restart.
return taker.schedule[taker.schedule_index][2] = len(taker.honest_makers)
retry_str = "Retrying with: " + str(taker.schedule[
taker.schedule_index][2]) + " counterparties."
tumble_log.info(retry_str)
log.info(retry_str)
taker.set_honest_only(True)
taker.schedule_index -= 1
return
#There were not enough honest counterparties. #There were not enough honest counterparties.
#Tumbler is aggressive in trying to complete; we tweak the schedule #Tumbler is aggressive in trying to complete; we tweak the schedule

Write Preview
Loading…
Cancel
Save