V12 Release

akhavr

Active Member
Oct 11, 2014
817
408
133
Nope, can't find the Ukraine on the charts, localbitcoins must have the data though, perhaps theses guys just haven't got round to adding you yet,
http://bitcoincharts.com/markets/list/
I can count the number of people that does face-to-face BTC trading in Ukraine regularly using my fingers. Not using my legs. And I'll still have some fingers left for new candidates :)

That's why creating a clone of localbitcoins for a less popular currency will work only for hightech hubs like SV. Maybe London and Berlin (didn't visit, can't say)
 
  • Like
Reactions: moli

Solarminer

Well-known Member
Apr 4, 2015
762
922
163
They don't issue card themselves, only banks can do it. Fiat funds aren't "on cards", they're on bank accounts, which may or may not be tied to those cards.
So getting cards doesn't help much if you can't have either of
- a fiat bank account
- a company that allows you to operate its subaccount

One interesting approach I'm looking for is having a cryptotoken pegged 1:1 to a fiat currency (e.g. BitUSD, NuBits), but it still requires some kind of an interface to a SWIFT/SEPA/ACH/card processing. Yet.
Can you make it work with a debt based account like a visa/mastercard? Then just use an outside service to settle the account after x transactions. You would need a party to use for a credit line but does that have to be a bank?
 

akhavr

Active Member
Oct 11, 2014
817
408
133
Can you make it work with a debt based account like a visa/mastercard? Then just use an outside service to settle the account after x transactions. You would need a party to use for a credit line but does that have to be a bank?
Who will hold a funds on a debit account, that has this card tied to?
Regulators view this person/organization as a custodian and regulate them (for a reason).
 

fuzzyduck

Active Member
Feb 19, 2015
137
113
93
Im trying to get .51 to run for over 3 afterwork evenings of trying now.... Still 'not capable masternode: Could not connect to xxx.xxx.xxx.xxx:9999'

tried:
-compiled from source like i always do. (raspberry pi 2)
-running wallet .51 to try to start it
-check port on router, 9999
-deleted all files on masternode except dash.conf for a full rebuild.
-deleted all files on wallet except the wallet and conf, for a full rebuild
-sweared at it
-drank booze to ease myself

Also. I think bootstrap is bugged (at least for me). After about block 30.000-ish it stops syncing and renames the bootstrap.dat to bootstrap.dat.old. Restarting dashd continues the sync. Maybe thats the key. Ill try above stuff, including the booze, excluding the swearing and excluding the bootstrap one more time tonight and report back.
To follow up on my own problem. My log contains mainly:


2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 UpdateTip: new best=00000000003d80fea2d93bde78cba94bfb6a30afee58cf8991e0216bef607822 height=25342 log2_work=52.342203 tx=97753 date=2014-02-28 12:20:23 progress=0.026413 cache=31464
2015-09-01 19:08:21 ProcessNewBlock : ACCEPTED
2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:21 UpdateTip: new best=00000000007b32d9a6d51e55474f450f185ceec5f28554af893e7b2d08cc81f0 height=25343 log2_work=52.342364 tx=97754 date=2014-02-28 12:20:27 progress=0.026413 cache=31465
2015-09-01 19:08:21 ProcessNewBlock : ACCEPTED
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 ProcessNewBlock : ACCEPTED
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 ProcessNewBlock : ACCEPTED
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 CheckBlock() : skipping transaction locking checks
2015-09-01 19:08:26 ProcessNewBlock : ACCEPTED

-Besides although i removed maxconnections, its stuck at exactly 8 connections always.
-During startup of dashd it reads about 7000 blocks, then it renames it to bootstrap.dat.OLD and starts downloading them.
-Block count creeps up sloooowwwwly.
-I got this also:

Binding RPC on address :: port 9998 (IPv4+IPv6 bind any: 1)
2015-09-01 18:02:15 ERROR: Binding RPC on address :: port 9998 failed: open: Address family not supported by protocol
2015-09-01 18:02:15 Binding RPC on address 0.0.0.0 port 9998 (IPv4+IPv6 bind any: 1)

and

Error: Couldn't open socket for incoming connections (socket returned error Address family not supported by protocol (97))
2015-09-01 18:02:15 Bound to 0.0.0.0:9999


EDIT: am i hijacking the thread? If so sorry... :p
EDIT2: Hold on. its adding blocks like the wind now. Boostrap.dat still live and not renamed.. Im so confused.
EDIT3: Ok stopped...

tail -f debug.log
2015-09-01 19:22:38 keypool reserve 2
2015-09-01 19:22:38 keypool return 2
2015-09-01 19:22:38 keypool reserve 2
2015-09-01 19:22:38 keypool return 2
2015-09-01 19:22:40 keypool reserve 2
2015-09-01 19:22:40 keypool return 2
2015-09-01 19:23:01 keypool reserve 2
2015-09-01 19:23:01 keypool return 2
2015-09-01 19:25:39 keypool reserve 2
2015-09-01 19:25:39 keypool return 2
2015-09-01 19:27:35 Loaded 24810 blocks from external file in 899974ms

renamed to .Old again :(
Back to 'skipping transaction locking checks'

EDIT4: concluded that 'skipping transaction locking checks' is in fact importing blocks from file. When using 'tail -f debug.log' all this stuff flies by on screen.
 
Last edited by a moderator:

moli

Grizzled Member
Aug 5, 2014
3,255
1,830
1,183
My wallets have been getting more and more "Synchronization failed...", and it took "mnsync reset" 6 times for one wallet to sync through.
This is v0.12.0.51-afa37a8 - win64-bit.


upload_2015-9-1_16-9-11.png
 

Solarminer

Well-known Member
Apr 4, 2015
762
922
163
My wallets have been getting more and more "Synchronization failed...", and it took "mnsync reset" 6 times for one wallet to sync through.
This is v0.12.0.51-afa37a8 - win64-bit.
Same here. win64 windows 8. Gets to 75% and can take 10minutes to get through. Has failed a few times. It seems that this is depended on when it is syncing. Last night it couldn't get past 75%, even restarted several times. I didn't try the mnsync command, but did delete the peers.dat which didn't help Just tried loading now and it loaded right up in about a minute. I have an i7 and fast internet. The wait time is due to a network/communication failure - my machine isn't using the cpu or bandwidth while it is waiting.
 
  • Like
Reactions: moli

Solarminer

Well-known Member
Apr 4, 2015
762
922
163
Who will hold a funds on a debit account, that has this card tied to?
Regulators view this person/organization as a custodian and regulate them (for a reason).
Gotcha. It is always the fiat storage/transfer where the problems happen.
 
  • Like
Reactions: akhavr

Lukas_Jackson

Member
Nov 9, 2014
160
70
88
Im trying to get .51 to run for over 3 afterwork evenings of trying now.... Still 'not capable masternode: Could not connect to xxx.xxx.xxx.xxx:9999'
Sometimes it stuck at 50% or 75% mn winners syncing. Try on hot side './dash-cli mnsync reset', wait few minutes and then cold start. It helped me

edit: now I see that you tried :)
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
My masternode's run-time is 6 days and has not been paid for 11 days. I keep seeing masternodes that have run-times of like 14 days being chosen for payment (having been paid only 4 days ago) instead of my MN. For instance this one...

XnG1dkS8EV2pMPTk6JDE4AvTesQa4a3FjT

How is it possible that this masternode's running time is 14 days? Didn't they have to restart when they upgraded? Is the payment system working as designed? Is it being gamed?
Masternodes that upgraded when v12 came out may never have gone off line. It's been 3 weeks now. The account you posted: XnG1dkS8EV2pMPTk6JDE4AvTesQa4a3FjT, does not show up on dashninja.pl. Something is wrong with it. If you're not running version 12, you will not get paid. You may also have gotten on the wrong fork (currently, we're on block number 329369 https://chainz.cryptoid.info/dash/ )

Now I realize you asked this question a while ago, and you may have taken hat MN down in the mean time, but just in case:
make sure you're running the latest version
make sure that you reindexed your node. Personally, I remove everything in my .dash folder EXCEPT wallet.dat and my .conf files (dash.conf and / or masternode.conf)
make sure, after the daemon has updated completely, start your local wallet with the 1000 dash in it, and start your masternode, check that your remote sees it with masternode debug, then it should be fine.

Anytime we are on a protocol that isn't changing (usually only changes with new versions), then you should never restart the masternode from the local wallet. That will send you to the back of the line payment wise. As long as your masternode is not off line more than 70 minutes, all you need to do is restart your remote daemon.

If you haven't already, check the wiki, there are instructions on how to implement a cron job to restart the daemon should your server be restarted, which sometimes happens with service providers. Also, there is now a script to check if your masternode is running, ever few minutes, and if it's not, it restarts the daemon. These two backups are really a good idea to implement as you can't be there 24-7.
 

Geert

Member
Aug 26, 2015
259
82
88
Masternodes that upgraded when v12 came out may never have gone off line...
Thank you for responding! First off, dashninja is a little wonky lately -- I usually have to load the page that details the MN, and then refresh the page to get it to display anything. Dashnodes works better...

I'm using a hosting service, so I am at their mercy if my MN goes offline and I have to do a remote restart.

I'm starting to wonder if it's really possible to create a mechanism whereby the MNs reliably come to a consensus about who among them gets rewarded next (currently we can't seem to maintain consensus). This is a hard problem that probably needs to be modeled using chaos theory (not that I know much about it). It might feature unexpected oscillations and feedback loops. Also, as the software is open source, what about people tweaking it to try to gain advantage. For example, is it possible to alter the code such that one could do a remote restart on the sly? If Dash gets more valuable, you can be sure people will try to game this.

Thanks again.
 
Last edited by a moderator:

camosoul

Grizzled Member
Sep 19, 2014
2,261
1,130
1,183
Do we have some development plan that will promote the trading of DASH with fiat? I think the MN and InstantX features are great but we really need some good trading volumes to compete with other cryptocurrencies. Does the profit from MN making margin trading unlikely because we don't really have a swap market? I mean if you have thousands of DASH, would you run MNs rather than offering swaps? I think the leverage by taking out a swap is what makes trading so profitable. We might not need to think about this when large exchange like Bitfinex had DASH/BTC and DASH/USD pair. Now, it seems relevant to us. I really don't want to go to Cryptsy for DASH trading. I don't know why, I just don't feel comfortable with Cryptsy.
The thing holding DASH back is understanding...

When Silk Road first launched, the guv and their collusive media pals put in motion a large misinformation campaign to define Bitcon as "Anonymous Internet Devil Money for teh DRUGGGZZZZ!" They emphasized the "anonymous" part to get the fools that didn't know otherwise to make low-hanging fruit of themselves. To this day, the key to catching most of them has still been "follow the money." They make no effort to cover their tracks because the falsely believe they are not leaving any.

This false belief that there is any kind of privacy at all with Bitcoin and all of it's clones is still believed by many. They don't understand why DASH exists. They have so much pride in the coin they hold that any attempt to educate them is considered trolling or promoting a pump/dump of your/their coin of choice...

The key is finding a way to educate them in a way that doesn't seem like you're attacking their precious shitclone. And they can't really be blamed too much, with the huge amount of BS in crypto, even a mature person would be skeptical and defensive to the point that they may simply refuse to look around... Most traders have no clue how DASH works, or any other coin for that matter. DASH is unattractive for trading because the masternodes' proof of service holdings make it hard to manipulate, so there isn't a bunch of BS volume and big waves to make or lose money on.

No one has yet discovered the holy grail of how to educate an overly defensive, emotionally charged, cryptotard that their coin doesn't do what they think it does, without setting off their perfectly sensible defense mechanisms created as a result of the endless scamming and trolling inherent to crypto...

Further, DASH will never be a truly high volume trade because the senseless fluctuations that other coins have are pretty much eliminated, and pump/dump schemes are too expensive compared to what they could accomplish in any other coin for the same investment. In a market that primarily exists so that cryptotards can force each other to trade empty bags, DASH has kind of alienated itself by being more inherently stable than the competitors. The feature became a bug due to the userbase of crypto in general being a bunch of fuckwipes.

I believe DASH's future does not lie on the exchanges. It's inherently more stable and has instant TXes... It just won't ever live on the exchanges because it's not wildly fluctuating and anything else is a more effective/cheaper manipulation target. DASH is a different beast and it treads into territory that most cryptos forgot to become as they all huddled around the Ouija Board Chart abbreviation-trading crack pipe. They forgot to be cryptoCURRENCIES and a cult of derp has evolved around that failing. DASH has remembered to be a cryptoCURRENCY, but it's just wandering off into the cold darkness because everyone is addicted to the unintended crack-pipe of speculative trading on nonsense fluctuations and manipulation that the rest of crypto has become.

Education and promotion are key. The differentiating "it's actually a currency" features are already there. But nobody knows why. They think it's a solution to a problem that doesn't exist because they fell for the initial propaganda barrage and never looked back. Finding someone who knows what crypto is, is rare. Finding someone who knows that Bitcoin (and all of it's clones) is not anonymous already, is even more rare... I don't know what the Holy Grail is for bridging that gulf of cluelessness without setting off their alarms. That's what's really holding it back... Then, whenever that Holy Grail is discovered, it won't be exchanges, but real-world use via IX, that accelerates adoption. The stability is a GOOD THING for an actual CURRENCY, but the traders, and thus the exchanges that make commissions off of them, don't think so because their profit is centered around instability of a speculative and manipulative investment, which is what the bulk of crypto, and the mentality around it, has become. DASH's innovations has made it the "straight edge" in a community of crack-heads. DASH is the odd one out in a community of so-called "people" who have no idea why a person would care about anything but jonsing for their next rock.

As I've said before, the smart money is already in DASH. There's just not that much smart money...
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
Thank you for responding! First off, dashninja is a little wonky lately -- I usually have to load the page that details the MN, and then refresh the page to get it to display anything. Dashnodes works better...

I'm using a hosting service, so I am at their mercy if my MN goes offline and I have to do a remote restart.

I'm starting to wonder if it's really possible to create a mechanism whereby the MNs reliably come to a consensus about who among them gets rewarded next (currently we can't seem to maintain consensus). This is a hard problem that probably needs to be modeled using chaos theory (not that I know much about it). It might feature unexpected oscillations and feedback loops. Also, as the software is open source, what about people tweaking it to try to gain advantage. For example, is it possible to alter the code such that one could do a remote restart on the sly? If Dash gets more valuable, you can be sure people will try to game this.

Thanks again.
Dashninja does have a lot of scripts running the pages, and I also have trouble keeping it open on my old machine :) But try this: https://dashpay.atlassian.net/wiki/pages/viewpage.action?pageId=8880256 which will monitor your masternode to be sure it's running, and restart it if it should stop. actually, that script is more complex, and is supposed to deliver an email notification. All you really need is to restart. See how I have it below :)

Also, So that if the server goes down, but comes back up within an hour, your masternode will be restarted automatically. Make a cron-job. You can do this either as root or as the user. I think it's better as the user of the masternode, so a cron job per masternode/user. You do this so:

type in:
crontab -e

choose your favorite editing thingy, I use nano, others vi (most popular) and hopefully one of them is installed on your machine. Just enter the number next to the choice:

2

At the bottom of the file that opens, insert this: (Use the down arrow key to get to the bottom of the file)

Copy and past into screen:

@reboot /usr/local/bin/dashd
*/1 * * * * /home/ingie/mn_watch.sh >/dev/null 2>&1


Control o writes the changes
Control x saves the file and closes nano

Well, that's what I have in there. The first line means to restart dashd when system is rebooted. The second line checks to see if dashd is running, and if not, executes the mn_watch.sh script I placed in home/user/ folder. The second cron job on the list checks every minute (probably too often, come to think of it) and if dashd isn't running, executes this sh file that I have at that location. It's called mn_watch.sh, I made it in nano, and the only thing inside is this:

To make this file, again open nano:

nano mn_watch.sh

this opens a new file named mn_watch.sh.

Now copy this and paste it into your screen:

#!/bin/bash
#mn_watch.sh
#make sure a process is always running.

process=dashd
makerun="/usr/local/bin/dashd" #where my dashd, dash-cli and dash-tx reside

if ps ax | grep -v grep | grep $[d]ashd > /dev/null
then
exit
else
$makerun &
fi


exit

(or better, the one at the bottom of this comment)

and finally, again, do control o and control x to save.
And this is because I keep my dashd, dash-cli and dash-tx files in usr/local/bin so that I don't have to do ./ before every command. Also, I have two masternodes on this machine, and I only have to update once, then start dashd for each user.

explaination: 'pidof dashd' refers to dash.pid which you can see in your .dash folder when dashd is running. I think it links to the running service. So if it's not there, it means dashd isn't running, hence the "then" statement of executing dashd. make sense? I'm still learning too. LOL, I just tried to open my dash.pid file to see what's inside and it disappeared! Wonder if my dash will try to restart now, LOL.

I really think your masternodes might not be running properly. The consensus system has actually been running perfectly as I understand it?

Edit: My pid file version wasn't working, so I tried this and it seems to work.
Edit2 The original does indeed work, don't know what went wrong in my testing, but this is simpler, and probably better, as it uses pid file instead of grep :)

#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi
 
Last edited by a moderator:

elbereth

Active Member
Dash Support Group
Mar 25, 2014
461
484
133
Costa Rica
dashninja.pl
Dash Address
XkfkHqMnhvQovo7kXQjvnNiFnQhRNZYCsz
Thank you for responding! First off, dashninja is a little wonky lately -- I usually have to load the page that details the MN, and then refresh the page to get it to display anything.
I just fixed this. It was a typo in my code causing first call to the masternode end-point of the API to fail but result was cached successfully. Therefore when reloading it used the cache and was a success.
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
Half of my v0.12.0.51 MNs died. I think there are still some bugs in it. The machine I ran dashd has plenty of RAM and disk space. There was no sign of running memory or disk space. Anyone else having the same problem?
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
Further, DASH will never be a truly high volume trade because the senseless fluctuations that other coins have are pretty much eliminated, and pump/dump schemes are too expensive compared to what they could accomplish in any other coin for the same investment. In a market that primarily exists so that cryptotards can force each other to trade empty bags, DASH has kind of alienated itself by being more inherently stable than the competitors. The feature became a bug due to the userbase of crypto in general being a bunch of fuckwipes.
That's the dilemma. I like DASH because it is hard to manipulate when compared to other "shit"coins. But, this at the same time has alienated itself in the way you described.
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
Dashninja does have a lot of scripts running the pages, and I also have trouble keeping it open on my old machine :) But try this: https://dashpay.atlassian.net/wiki/pages/viewpage.action?pageId=8880256 which will monitor your masternode to be sure it's running, and restart it if it should stop. actually, that script is more complex, and is supposed to deliver an email notification. All you really need is to restart. See how I have it below :)

Also, So that if the server goes down, but comes back up within an hour, your masternode will be restarted automatically. Make a cron-job. You can do this either as root or as the user. I think it's better as the user of the masternode, so a cron job per masternode/user. You do this so:

type in:
crontab -e

choose your favorite editing thingy, I use nano, others vi (most popular) and hopefully one of them is installed on your machine. Just enter the number next to the choice:

2

At the bottom of the file that opens, insert this: (Use the down arrow key to get to the bottom of the file)

Copy and past into screen:

@reboot /usr/local/bin/dashd
*/1 * * * * /home/ingie/mn_watch.sh >/dev/null 2>&1


Control o writes the changes
Control x saves the file and closes nano

Well, that's what I have in there. The first line means to restart dashd when system is rebooted. The second line checks to see if dashd is running, and if not, executes the mn_watch.sh script I placed in home/user/ folder. The second cron job on the list checks every minute (probably too often, come to think of it) and if dashd isn't running, executes this sh file that I have at that location. It's called mn_watch.sh, I made it in nano, and the only thing inside is this:

To make this file, again open nano:

nano mn_watch.sh

this opens a new file named mn_watch.sh.

Now copy this and paste it into your screen:

#!/bin/bash
#mn_watch.sh
#make sure a process is always running.

process=dashd
makerun="/usr/local/bin/dashd" #where my dashd, dash-cli and dash-tx reside

if ps ax | grep -v grep | grep $[d]ashd > /dev/null
then
exit
else
$makerun &
fi


exit

(or better, the one at the bottom of this comment)

and finally, again, do control o and control x to save.
And this is because I keep my dashd, dash-cli and dash-tx files in usr/local/bin so that I don't have to do ./ before every command. Also, I have two masternodes on this machine, and I only have to update once, then start dashd for each user.

explaination: 'pidof dashd' refers to dash.pid which you can see in your .dash folder when dashd is running. I think it links to the running service. So if it's not there, it means dashd isn't running, hence the "then" statement of executing dashd. make sense? I'm still learning too. LOL, I just tried to open my dash.pid file to see what's inside and it disappeared! Wonder if my dash will try to restart now, LOL.

I really think your masternodes might not be running properly. The consensus system has actually been running perfectly as I understand it?

Edit: My pid file version wasn't working, so I tried this and it seems to work.
Edit2 The original does indeed work, don't know what went wrong in my testing, but this is simpler, and probably better, as it uses pid file instead of grep :)

#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi
There is a problem with this bash script in that, if you have two users running dashd on one server, if one dashd daemon goes down, it won't restart because it thinks dashd is still working as it's still on with the other user.

How can I have it check only if the local daemon for the user is running? Thanks for any help :)

#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi
 

Defacto

Member
Aug 20, 2015
50
30
58
Warsaw
Dash Address
Xj8PUS4r9vk5y8BvoovMoRpbhHXuW6uj5o
There is new v52 on github. It's not activate mn correctly. It can't be also activated remotly from cold walet. Something is wrong with this version.
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
There is a problem with this bash script in that, if you have two users running dashd on one server, if one dashd daemon goes down, it won't restart because it thinks dashd is still working as it's still on with the other user.

How can I have it check only if the local daemon for the user is running? Thanks for any help :)

#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi
I have a different setup so my crontab probably won't work for you. You may be able to use the following system crontab entries. This is not tested and I wrote it by modifying my crontab according to your need. If you have any problem, please let us know whether it works for you or not.

Code:
*/5 * * * * user1 if [ -z $(fuser -c ~user1/.dash 2>&-) ] ; then sleep $(jot -r 1 0 299) ; dashd 1>&- ; fi
*/5 * * * * user2 if [ -z $(fuser -c ~user2/.dash 2>&-) ] ; then sleep $(jot -r 1 0 299) ; dashd 1>&- ; fi
A system crontab (i.e. crontab by root) can specify the user to run a command. So, change user1 and user2 accordingly. One single line for one MN.

That's sad. We had MN running for months without crashes with the last v0.11 release. And, I actually deleted my old script to restart a crashed MN. Hope v0.12.0.52 will resolve these bugs.
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
There is new v52 on github. It's not activate mn correctly. It can't be also activated remotly from cold walet. Something is wrong with this version.
Good to know that the developers are still working on v0.12.0.x. There are definitely bugs to be resolved before one can run this for months without crashing.
 
  • Like
Reactions: moli

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
Thanks, GermanRed+, but I think the problem is in my bash script. I just discovered another way that is supposed to be better:

until dashd;do
echo "Server 'dashd' crashed with exit code $?. Respawning..">&2
sleep 1
done


Only thing is, to test it, I need to crash dashd, because if I stop it without error, it will stay stopped (because I would have asked it to, and it would have exit code 0 (I believe?)

So, question is, anyone know how to crash dashd? LOL Maybe pkill?

Damn, it's not working.
 

flare

Administrator
Dash Core Team
Moderator
May 18, 2014
2,286
2,404
1,183
Germany
There is new v52 on github. It's not activate mn correctly. It can't be also activated remotly from cold walet. Something is wrong with this version.
There was a accidental merge, I've reset master to v51. Please update your local git clone accordingly.
 
  • Like
Reactions: Defacto

GermanRed+

Active Member
Aug 28, 2014
299
109
113
Thanks, GermanRed+, but I think the problem is in my bash script. I just discovered another way that is supposed to be better:

until dashd;do
echo "Server 'dashd' crashed with exit code $?. Respawning..">&2
sleep 1
done


Only thing is, to test it, I need to crash dashd, because if I stop it without error, it will stay stopped (because I would have asked it to, and it would have exit code 0 (I believe?)

So, question is, anyone know how to crash dashd? LOL Maybe pkill?

Damn, it's not working.
That won't loop forever. If your dashd return 0, the loop will break and continue after the line done.
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
That won't loop forever. If your dashd return 0, the loop will break and continue after the line done.
Ugh, it's late here, and I've been reading and reading all kinds of ways to do this. Thank you for your comment, but could your please tell me if the script above actually restarts dashd if it stops running?

Or is there a way to check to see if user1 instance of dash has stopped and needs a restart. I think what is happening in the other script:
#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi

is that it sees dashd running on the other user's login, so it won't restart the first user's daemon. Probably the same problem with this new looping idea.

in the above, I've tried to find out what the -z is, but I can't find anything. I assume it's shorthand for "not running"?

And with the looping script, after pkilling the daemon, it didn't restart after a few minutes (cron set to 1 minute) so I just executed the sh file, and it started the daemon. I can't figure out what is going wrong here, ugh! LOL

Thank you again for your help :)
 

GermanRed+

Active Member
Aug 28, 2014
299
109
113
Ugh, it's late here, and I've been reading and reading all kinds of ways to do this. Thank you for your comment, but could your please tell me if the script above actually restarts dashd if it stops running?

Or is there a way to check to see if user1 instance of dash has stopped and needs a restart. I think what is happening in the other script:
#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi

is that it sees dashd running on the other user's login, so it won't restart the first user's daemon. Probably the same problem with this new looping idea.

in the above, I've tried to find out what the -z is, but I can't find anything. I assume it's shorthand for "not running"?

And with the looping script, after pkilling the daemon, it didn't restart after a few minutes (cron set to 1 minute) so I just executed the sh file, and it started the daemon. I can't figure out what is going wrong here, ugh! LOL

Thank you again for your help :)
It will only restart it once. And then, the script will exit unless you somehow go back and loop again. The bigger problem with that script is that you run dashd every second when your dashd is running just fine. You do not want to do that at all. I don't even want to run fuser every 5 minutes because it makes system call. That's why I think the developer should fix the bugs so that we don't need to run some check in some infinite loop.
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
Well, I guess you can't trust anyone on the internet, LOL. I don't understand why I can't use

#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi


when the daemon is running on another user's account? ugh!
 

UdjinM6

Official Dash Dev
Core Developer
Dash Core Team
May 20, 2014
3,639
3,537
1,183
Ugh, it's late here, and I've been reading and reading all kinds of ways to do this. Thank you for your comment, but could your please tell me if the script above actually restarts dashd if it stops running?

Or is there a way to check to see if user1 instance of dash has stopped and needs a restart. I think what is happening in the other script:
#!/bin/bash
if [ -z `pidof dashd` ]; then
/usr/local/bin/dashd
fi

is that it sees dashd running on the other user's login, so it won't restart the first user's daemon. Probably the same problem with this new looping idea.

in the above, I've tried to find out what the -z is, but I can't find anything. I assume it's shorthand for "not running"?

And with the looping script, after pkilling the daemon, it didn't restart after a few minutes (cron set to 1 minute) so I just executed the sh file, and it started the daemon. I can't figure out what is going wrong here, ugh! LOL

Thank you again for your help :)
True if the length of "STRING" is zero.
http://tldp.org/LDP/Bash-Beginners-Guide/html/sect_07_01.html

Half of my v0.12.0.51 MNs died. I think there are still some bugs in it. The machine I ran dashd has plenty of RAM and disk space. There was no sign of running memory or disk space. Anyone else having the same problem?
Is dashd mentioned in /var/log/syslog?
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
UdjinM6 Ah! So if they can't find the string "dashd" it is zero, and off, then it executes my script, eh? Thank you for that! I just don't know the words to search for!
 

UdjinM6

Official Dash Dev
Core Developer
Dash Core Team
May 20, 2014
3,639
3,537
1,183
UdjinM6 Ah! So if they can't find the string "dashd" it is zero, and off, then it executes my script, eh? Thank you for that! I just don't know the words to search for!
Almost ;) "pidof someProcess" gives you pid of someProcess if it's running or an empty string if nothing was found
 

TanteStefana

Grizzled Member
Foundation Member
Mar 9, 2014
2,871
1,863
1,283
Is there a way to do pidof someprocess by someuser? Its the only thing I can think of that is keeping this from working when one of the users has a daemon running and the other doesn't. Thanks for the help, UdjinM6!!