HELP(!) - Impossible PoW recipient address msg

[chan] bitmessage
Jul 14 09:33

HELP(!) - Impossible PoW recipient address msg Tough Day, broken sleep. Zombie assistance request. Even the Nero double strength coffee shots aint workin' Sent a bunch of (short) messages to my home target system with ( I thought ) a different range of PoW private key settings. Starting with the lowest. Or so i thought. Screwed up - badly - all were created with an insane default Subsequent edited values are not the ones presented for first use by a sender elsewhere on the network. No surprise there. Is it possible to force propagation of the corrected PoW values in keys.dat over the network? If so - How? I can probably get remote shell access to the target system from here. Otherwise I'll have to reschedule the trial audience I'd planned to assemble later today. Operating from a laptop, away from home base. This BM is a daemon with no Qt GUI capability in the OS. Just a demo system really. Uses bmwrapper with a conventional email client. I may have some command line API tools somewhere but cant find 'em. Or remember the syntax and API commands to kill the queued items. It would be nice to send an RFC cancel email but i doubt that would work. How do I kill this queue - quickly - to avoid thermal issues with the laptoop - after restarting the daemon? It isnt in a VM so i cant throttle it to constrain resource usage. ( Sending this from the rather splendid BM GUI in a VM ) Anone else ever been in this situation - playing with PoW settings? Last recovered situations likethis 2 1/2 Years ago. But not remotely. Some people never learn... Helpppppp ;-) <*sigh*>

[chan] bitmessage
Jul 14 09:34


[chan] bitmessage
Jul 16 16:40

Shut down bitmessage then kill any remaining python process. use sqlite3 to delete the offending messages, and delete all rows in the object processor queue table. You might lose some data, but it will fix the problem.

Jul 16 16:40

If it is ok for the queued messages to stay in the send folder but be ignored, the easiest workaround is to reduce maxacceptablenoncetrialsperbyte and maxacceptablepayloadlengthextrabytes in keys.dat. A value of 10000 is reasonable on a low end machine. Shut down PyBitmessage prior to editing the file. You may have to kill the process forcibly if it's stuck in the PoW. Upon restarting, it won't do PoW for those messages, and if you want you can use RPC to delete them. This is probably the fastest way to resolve your issue. While it is in theory possible to update your pubkey metadata, it's complicated and I'm not aware of an existing description of the process (I have a vague idea how it could work but I never tried it and would have to do an investigation). I've known for a while that this behaves non-obviously and there is potential for improvement but it's very rare and not a high priority. Peter Surda Bitmessage core developer

[chan] bitmessage
Jul 16 20:44

Using the GUI it is possible to delete outgoing messages regardless of their state. If PoW has not yet been started on a message then obviously it won't ever start but if PoW is currently being performed on a now deleted message then a simple restart of bitmessage should stop the PoW. Hopefully it should be quite similar with the API - just delete the messages as normal (getSentMessagesByAddress, trashSentMessage). I'm not sure how to go about restarting bitmessage remotely in the current situation though. However in the future in lieu of an API function to call, you can set up the API to run a script when API events occur. Just set apinotifypath in keys.dat to the desired script. This functionality is quite basic, there are only 2 events startingUp, and newMessage which are the arguments passed to the script. The script (which you would have to write) when called by the API would in turn connect to the API to try to retrieve a message from a particular sender, process the message for actions to take, and carry out those actions. An example action would be to send SIGINT or SIGTERM (these are supposed to work when bitmessage is in daemon mode) to bitmessage to get it to shutdown normally, then just start bitmessage again.

[chan] bitmessage
Jul 16 20:53

MANY Thanks Peter. I had actually compounded the stupidity - On closer inspection of those two magic numbers. For equipment with plenty of DRAM PyBM is almost always running from within a VM. But not on my "worst case" 7+ YO Dell laptop. Assigning resources to keep max temp of equipment in bounds is easy. Had not correctly changed ALL of the values from a daemon keys.dat inside a large Dev Workstation VM - deliberately running High PoW comms tests as an entropy feed for exotic key generation on the host. When it was transplanted to the Laptop host and edited ages ago. All is now in Harmony. Acceptable ratios of steam :- coffee cup(s) / ears / laptop now in effect. I did look, briefly, at a way to throttle PoW c/o some for of "listener" communicating with the daemon code, rather than an API extension. Some time ago. What would be the least invasive / spec changing code based way to do that? With physical access to the machine a magic key combo, held down to throttle "Up" and "Down" should be relatively easy. Wait states built into the compiled C PoW routine perhaps? Take this as a feature request; I broke a messages.dat file beyond all hope of repair using kill -9 five months ago. An "All or nothing" option for native Host running stuff is not good. ( daemons when the host also lacks Qt display capability to run the GUI, and/or large batches of messages need to be processed efficiently ). Needed while the exact private key demanded PoW network served dynamics are a mystery that people WILL find edge cases to fall off. And are obliged to run PyBM native on the Host because of insufficient resources to operate a VM solution. Earlier this morning I was standing a cliff edge. Good to have a push in the right direction of a decisive step ;-) Have a good afternoon Matey! You have significantly improved mine. Ta Muchly.

[chan] bitmessage

Subject Last Count
Building instructions for Windows Sep 24 17:53 13 is being cached by google Sep 24 16:51 1
How to.. Sep 24 16:43 2
BitText XHKhFPCDzj: ultimate bitmessage forum Sep 24 16:36 1
BitText Error Sep 24 16:32 1
BitText LIST Sep 24 16:16 2
HELP Sep 24 15:49 2
BitText ADD confirmation Sep 24 15:34 1
TTL Tweaks Sep 24 15:27 12
turn MiNode into a full client - simple task ! Sep 24 15:12 1
use MiNode py3 app to route a BM via "stream 7" Sep 24 14:07 1
BM GUI via API using monkey studio Sep 23 20:35 26
BM GUI via API using KDevelop Sep 23 18:21 1
Email client integration Sep 23 16:34 10
got green light Sep 23 10:59 3
pyBM eats too many CPU cycles Sep 23 10:36 22
Git pull error Sep 23 08:49 17
What's a status of DevTalk pseudo-mailing list? Sep 23 07:44 7
Kdevelop + qt-designer for python Sep 22 21:47 1
NATO member Turkey boast that Russian S-400 SAMs can take out American B-52s, F-22s and Tomahawks Sep 22 12:24 1
bug? pyBM eats 17% of my 6core CPU cycles SOLVED :-) Sep 22 11:41 1
bug? pyBM eats 17% of my 6core CPU cycles Sep 22 11:37 21
monkey studio Sep 22 11:06 1 Sep 22 10:56 2
so stream 7 cannot work ? Then what did I do ? Sep 22 10:44 1
How make new stream? Sep 22 10:29 22
(Qt4 security staus?) Sep 22 10:27 6
Question Sep 22 05:49 8
GUI choice is Good. Ncurses anyone? Sep 22 05:35 6
attitude Sep 22 03:21 2
(Qt4 security status?) Sep 22 02:57 2
Searching for BM address for g1itch/Dmitri Bogomolov Sep 21 15:41 4
QT designer and Monkey GUI builder for pyBM Sep 21 15:39 1
bug report: Connected hosts: 0 -- SOLVED :-) retracted Sep 21 13:35 1
bug report: Connected hosts: 0 -- SOLVED :-) Sep 21 13:32 2
Gabon has teamed up with militant conservation group Sea Shepherd Sep 21 13:17 1
twister micro blogging Sep 21 12:40 6
rating in pybitmessage snapshot build Sep 21 12:38 3
#Assange Sep 21 11:45 4
bug report: Connected hosts: 0 Sep 21 11:09 7
How would you describe the technology behind Bitmessage Sep 21 09:43 4
does anyone get BM over tor to work ? SOLVED :-) Sep 20 22:03 1
does anyone get BM over tor to work ? Sep 20 22:00 12
what needs to be in a real BM GUI via API - web Sep 20 21:55 20
what needs to be in a real BM GUI via API Sep 20 21:52 14
BM GUI via API Sep 20 21:47 1
bot bug: bot keeps sending me a list - ACK is missing Sep 20 21:38 2
get BM over tor to work ( see UBF ) Sep 20 21:32 1
GUI Sep 20 20:38 1
BM future & features Sep 20 16:41 7
Feature Mashup Sep 20 16:40 1
BitMessage Secure Station (Developer Version) open-core open-hardware project PCB routing advancement : 75% Sep 20 16:01 9
py typo-squatting Sep 20 15:41 1
does anyone get BM over tor to work ?? --> poss. solution now? Sep 20 12:09 2
does anyone get BM over tor to work ?? Sep 20 11:17 5
what needs to be in a real BM GUI Sep 20 11:14 27
Is anyone there? Sep 20 07:47 5
Rewarded help needed for scanning old amazon (Or any other website) shipment barcodes (Reward $10 in Bitcoin per scan). Sep 19 22:25 1
Bitmessage noise script Sep 19 00:48 15
BitMessage Secure Station's architecture security review : White Papers & Publications about Designing Secure Hardware and fighting Hardware Backdoors. Sep 19 00:48 4
How to compliment a female cryptographer? (joke) Sep 19 00:47 3
Great article about fighting Hardware Backdoors and how to design secure open-hardware open-core systems (BitMessage Secure Station) Sep 17 11:58 1
Question about curves Sep 16 21:27 10
Streams Sep 16 10:39 4
BitMessage Secure Station open-core open-hardware project news. Sep 15 22:07 8
Tor Browser Sandbox Sep 13 21:39 1
BitMessage Secure Station "version developer" security update Sep 13 20:56 1
ISS Space Station - Augmented Virtual Reality Sep 13 12:41 2
TTL of pubkey requests Sep 13 05:52 2
encryption in BM Sep 12 16:55 20
BM connection security Sep 12 08:26 8
First pictures of the in routing PCB of the BitMessage Secure Station, version developper (Implemented with a PIC 24 Microcontroller, and not a Spartan 6 FPGA) Sep 11 13:37 1
Another Question about curves Sep 11 09:49 3
onion issues/questions? Sep 11 06:53 32
bitmessage has forward secrecy? Sep 9 08:35 17
mixmaster / mixminion routing in bitmessage Sep 9 08:23 6
Passphrase encryptsion for keys.dat? Sep 8 20:33 14
Bitmessage Addresses Question Sep 8 09:46 2
Anatomy of encrypted payload Sep 6 08:24 8
BM / Sep 6 07:07 6
Bitmessage Mix Routing Sep 6 03:32 1
Unencrypted image Sep 5 20:22 5
Do PoW but don't send the object Sep 5 09:05 2
Question on BM source code Sep 5 08:42 3
What does this mean? Sep 2 05:03 3
Anybody willing to share their xmpp IDs? Sep 1 15:05 1
Short Addresses Sep 1 01:40 35
Bitmessage and Namecoin Aug 31 06:45 3