hello

how to connect to daemon behind router with only open port 698

Tell your router to send 698 to the host running monerod, presumably on 18080 (or 18081 if you want RPC).

You won't be able to connect to more than one server though.

if i dont have access tou router nothing can be done?

@moneromoo: Are there some issues with open public nodes? xmr.to , hashvault.pro etc nodes seem to be down... Is this still our sibling guy?

ew, not sibling

I don't run any of those.

Yea, I know. Maybe they didnt apply the ban list. What do you mean by ew? ;)

Hmm, that's the kind of thing you'd say if you step in shit.

Or anything else really gross.

Anyway. Might be DoSed for all I know. You'd have to ask those people to check logs.

Ok, I thought ew is one of these fancy abbreviations :-D

@moneromooo daemon to daemon not wallet to daemon

Hmm... Strange that all these nodes have been pulled down... opennode and moneroworld is down as well.

xmrpow: monero-project/monero-gui #3140#issuecomment-706440354

try the nodes from this comment

xmrpow: monero.fail/?page=1

lists public nodes

xmrto onion is up

moneroworld is up but the their other one, uwillrunanodesoon is down and has been for a while

xmr.to node seems affected by the +2 bug

node.supportxmr.com:18081 should be fine, also the others I posted in this link

selsta: funny that monero.fail lists xmrto as being up and properly synced

last page

yea, it is synced

but the target height is wrong which monero.fail does not check

ok, thx

selsta: I had the sibling problem the day before yesterday and then added the banlist. Today my node failed again....

the attacker can add new IPs

ban list is temporary measure

I ryl looking forward to the new point release :)

try to update the ban list, if you still have issues post the output of sync_info

so that we can find the attackers IP address

we will tag the release today hopefully

ok, then im just going to wait.

updating the ban list should not take long and will most likely resolve your issue

what value did you set for out_peers?

I set out_peers to 500

Where can the list be found again?

if you set out_peers to 500 it will increase the chance of getting connected to an attacker by 40x

keeping it at default 12 is recommended

should be enough to solve your issue most likely, but still here is the latest version: gui.xmr.pm/files/block.txt

Isnt it better supporting more connections in general?

Thanks

I don’t know.

Afaik low out peers is better for decentralization, bitcoin core hardcodes it to 8

<xmrpow "Isnt it better supporting more c"> Out connections just cost the network more over the long term in theory.

<selsta "Afaik low out peers is better fo"> Yeah, that’s a great example.

having many in_peers means your node is helping the network, having many out_peers is not as beneficial

Yup 🙂

selsta: Oops, then I meant it the other way round. Thanks selsta! ;)

Then it makes sense :)

I always find it difficult to wrap my head around in / out peers

Me too :)

Then I am going to set my in_peers to 500 ;)

in_peers is unlimited by default

so no need to :)

ok

selsta: Are you sure that it is set to unlimited by default. My in_peer number is basically not growing.

?

something between 20-80 in peer should be normal

I do have 3

did you recently restart your node?

yep...

Maybe it takes some time

will take a while until peers get connected again

technically, uwillrunanodesoon is up, its just so overloaded that it doesn't respond within the timeout

i think its a bandwidth issue because system resources look fine. even though i thought i was payin for 1 gbps unmetered

it mighta been 100 unmetered

Has anybody ever done tests to see what the limits for txs per time period that the monero-wallet-cli can do?

er... I mean the rpc

vtnerd had some numbers in the asm patch PR IIRC.

asm patch?

The patch that added an assembler implementation of crypto ops to the wallet.

ah ok

thanks

You're referring to this here? monero-project/monero #6337

Yes.

Thanks moneromooo, I'd never have found that without you.

Hello

Im settint a monero node and now want to connect to it with monero-cli, is not working..

I get Error: wallet failed to connect to daemon

What are the default ports..?

18080 is p2p 18081 is RPC and 18089 is restricted RPC

Is monero-cli RPC ?

yes it uses RPC to talk to the daemon

or restricted RPC if it's remote

since full RPC lets you do stuff you don't want randos doing on your node

There's no default restricted RPC port, to be precise.

Would that be the command: ./monero-wallet-cli --daemon-address loopback:18081 --trusted-daemon

my bad, Ig I 18089 is just often used

monero-wallet-cli uses daemon RPC.

If loopback ends up correcly looked up, it's right, but it's also the default.

Are you sure your daemon is running, and is done syncing ?

The daemon is ./monerod right?

Yes.

I can turn on the wallet see balance, etc.. I cant mine though..

The daemon is running not sure if its done syncing..

"status" in monerod

you can run 'monerod status' to see but do unsynced nodes refuse connections?

Thats what monerod says: Sync data returned a new top block candidate: 2244517 -> 2246925 [Your node is 2408 blocks (3.3 days) behind]

They don't, but it might time out.

tru

And *some* RPC are rejected when not synced.

So wait till it's done.

right not being synced def would prevent mining

So am i not synced yet?

Also the your node blocks behind number is going up instead of down..

Like the number of blocks behinde is increasing instead of decreasing over time..

Status returns: Height: 2244517/2246926 (99.9%)

Ah, it might be stuck. Restart monerod with --log-level 1.

That'll give you more info on why it's not progressing.

Its unprogressing like going backwards..

well it's not losing blocks from what I see it's just that new ones are made every ~2 minutes

Looks like the same to me.

Orphaned block..

Ah ok, is the new ones that make my lag increase then?

But if new ones are created faster than my syncing I might never be fully synced then, right?

yeah. you should run it would log-level=1 and we can get more info

or '--log-level 1' rather

Yes i have..

E Block recognized as orphaned and rejected, id = <450919b06788d3d0043885d6b063409033a4cac06886be377b223c3752fceef4>, height 2244518, parent in alt 0, parent in main 0 (parent <19e0db157746e8a02869c96aa922e8ea501414731be699081f6db05936983569>, current top <cb6abc4af9da93eafc70e75673382a9dabbb21e47b6a59cf577b92bb0cfe13db>, chain height 2244517

Did my blockchain become corrupted? I read something about popping blocks?

Looks likely. Gimme a min.

That looks like a correct block. There should be an error before that, can you check ?

Ok, now its working.. Before it was looping and periodically giving the orphan block error.. But now seems unstuck and moving forward..

Yes, ok.

That's odd... Can you paste the entire log to paste.debian.net or similar ?

Oh, the prev top is not right in that message. That's why.

You likely were on a local fork, which took a bit to resolve.

The log might shed more light.

So some of the peers I was connected to where feeding me the wrong blockchain?

Maybe.

>3 days seems pretty far behind for an organic fork situation. I do wonder if it's our old friend the attacker

sounds suspect as hell to me

i'd redo the whole chain

nahh if you get synced and matching other well known nodes you good

I get this in red too: E Setting timer on a shut down object

That's fine.

I think it's fine, just a rare known problem, you seem to have "skipped" a block in download. I've seen that reported before, but it's very rare.

So when you are fully synced you are no blocks behind anymore?

So you could not add the new block because you were missing the one just before it. A little hole. It resolved once you fetched that missing block.

And thats why i cant connect with ./monero-cli to ./monerod yet?

Right. Then you'l constantly catch up every 2 minutes on average as new blocks are created.

Probably. I'll try to make the wallet feedback better.

Ok thanks Lyza moneromooo

I spent a few hours trying to work it out, and seeing the blocks behind number increase was doing my head in..

Did you restart monerod in that time ?

Yes i've closed it ctrs+c and restarted ./monerod a few times

Then maybe something else is odd since this should have reset the download queue.

Or maybe you were only connected to asshole nodes and they did not give you the data.

Still checking and the blocks behind is still increasing..

Can you paste the output of "sync_info" to some paste site ?

I was 2427 and then 2434

Im a bit concerned about the ips being out there..

I checked, it seems mine is not showing..

Im having trouble copy pasting the terminal output.. When i scroll down it unselects..

Exit monerod maybe.

Ok..

there's also the log file in your bitmonero folder

nm commands like 'sync_info' don't print to the log I think

but you can pull up monerod again when it's already running like, 'monerod sync_info' in the command prompt, it'll print the info then exit. then maybe full screen the term

Thats what was happening lastly: pastebin.com/ZeZ9ARw6

And thats what was happening previously when it seemed to error more: pastebin.com/hX7dy4Es

Dunno, part of me thinks if i just let it run ./monerod it will eventually sync but the other thinks that it might never sync meanwhile wasting resources..

Like now i restarted ./monerod, im getting the pattern from pastebin 2 , with orphaned blocks and the recurrent error red wording..

Mmm , also let me say i pruned the blockchain in case that has something to do with it..

You have a lot of known malicious peers there.

Pruning's fine.

What kind of person steals from their own community?

Download this: gui.xmr.pm/files/block.txt

Then run monerod with an extra: --ban-list block.txt

Malicious peers?

Was that targeted or they are just ther in the monero network connecting to whoever happens to turn up? I doubt it was targeted nobody knew i was startint to run a monero node, and I had no monero ports open so there was no tell..

There's a known dishonest asshole that's been barking at us for years. He's now trying to degrade network operation by running lots of nodes that behave badly.

Unlikely it's just for you.

hello

hi

Ok, I got those malicous nodes banned.. Blocks behind is still increasing though, at a far bigger rate than a block per 2 min..

Then I need to see the level 1 log from the start of your run. You pasted part of the log only.

At least the first verification error.

Probably that ban list is not comprehensive as they may be updating their ips to bypass it right?

the ban list is actively maintained but it's always possible there could be some that aren't on there, sure

to me this seems like a corrupted blockchain

How do I fix that? Cant believe i got to redownload the whole blockchain..Also how come it became corrupted?

but wait for mooo to look at the logs

out of interest, do you know who it is that's trying to poison the network with bad nodes

Yea I was wondering too, an ex developer?

what's that three year old reddit link referring to?

I read Fluffypony did a pump and dump..

lol

is your nick fluffypony due to mlp or something

rainbow dash best pony btw

jess: unfortunately not related to MLP, but I get that a lot :)

^^

re: who is trying to poison the network, it's a former developer of a piece of mining software

he tried to contribute to Monero itself, and even raised funds from the CCS, but then when delivery time came he had all sorts of strings attached via a weird license

which goes against merge request rules for Monero

Was it before 2017? I'm not familiar with that part of the story

so he got all butthurt and left to start a Monero fork, and instead of just quietly working on things he's spent the last few years attacking Monero

and then getting angry when nobody listens

sech1: 2018 I think

that's interesting I didn't know about the ccs/licensing stuff either

how u know that is he or anyone else

he told everyone it was him

made a dumb website and everything

lorian29: he's open about it

what is his gain

What was that CCS? Some wallet scan speedups?

<lorian29> I think it's more personal but he did fork the Monero code to create a competing currency so, there's that. seems like a dead project though.

He's probably paid for it. No one would waste so much of their life just for something so dumb.

sech1: yes

Ah, that one I remember

I didn't know it was funded, lol

it was basically just the ASM speedups that vtnerd eventually did

fluffypony: is this the fireice guy or whatever

yup

same dude spamming you on freenode?

Yes.

he's quite dedicated

so he start a fork alone and after quit and because of that is doing all this

He was also apparently stalking a monero contributor's brother on the phone :D

Damn. I shouldn't laugh about it, it's not nice.

wow psycho

he's pretty clearly unwell imo

Yeah, either that or paid.

If I were going to pay somebody to attack Monero I feel like I could do better lol

find somebody better I mean

For the tech side, sure.

maybe he'd do it for cheaper since he's motivated heh

I mean also personality wise he mostly makes himself look bad I feel like

I wonder what I've done. Full rebuild, and ccache ain't speeding things up...

moneromooo: yeah I agree, could entirely be externally motivated / paid

his aim isn't to break Monero technically

it's to cast doubt, to call individuals into question, and to fracture the community

i've got a bit of experience with trolls and i think the best advice i could give is pretend he doesn't exist

jess: we do

occasionally we have to acknowledge his existence when people ask questions, like now

yeah

fair enough

He's good at starting from a grain of truth to make up a mountain of lies though.

Could get a job a JTRIG.

at

If not already there ^_^

lol

hi

jcpham: ltns

huh

long time no see :-P

i'm a real boy!

play with my strings

lol

we hang out in secret channels

usermask were set invite only 10 years ago sorry

if I told you the channel name they'd kill me

i was able to sync the latetst monero blockchain as of friday that was neat

hello Hail_Spacecake

fluffypony our passports just expired :\

<Lyza> "seems like a dead project though." RYO's still kicking, not that I always agree with the tact used to engage with others.

I guess I should have said dying not literally dead

is the wallet polling the monero daemon for new transactions or does it make some soft or bidirectional connection where the daemon can push updates to the wallet?

The former.

thx

If you use 0mq, it might do the latter too.

Not sure it was done already, but it was planned.

whats 0mq

An RPC layer.

ohh wow looking at it now.

The wallet doesn't use it, so you;d have yto have your own wallet code.

Like mymonero. Which uses it.

thats why im asking

the daemon supports it?

Yes.

Well, 0mq. Pub/sub, not sure.

thanks selsta

Over 25,000 transactions in the last 24h, blowing past our previous ATH:

This is a bit too steep rise in transactions. Is someone playing with the network?

Potentially, hard to know TBH.

Could try to do fee fingerprinting or timing analysis to try and see if there is a clear batch.

We also are well past our previous yearly ATH in USD and climbing vs BTC so could be speculators rushing onto/off exchanges

We're past 2-year ATH

It was $140+ in July 2018 last time

Even better 🙂

So could be speculators + the steady organic growth we’ve seen lately.

But again, Monero makes it hard to determine 😅

are we being attacked again?

Via transaction spam? Or what do you mean?

Nothing way out of the ordinary that I can see, midipoet

Mostly just a steady, busy day in the mempool

I can at least say 2 of those past 24h transactions are mine :)

:D

but why would you ever say such a thing

When just one is yours.

btw moneromooo they both ended up in mempool instantly this time

using 0.17.1.5 + ban list

Of a miner ?

I checked at xmrchain.net

My previous notes on a flooding attack, if helpful to anyone else worried about the TX count rise:

I wonder when we'll see dynamic block size kicking in for the first time. Has it ever activated before?

Yes.

Spectacularly!

that fork bug?

Thought to be activated by the bytecoin team.

Yes.

that left hard-coded block hash in the source?

well, I mean first time for real. It'll be interesting to watch

Well, it also went up normally later on.

It never went up a *lot* though.

Getting pretty close TBH: localmonero.co/blocks/stats/block-medians

Lots of blocks hitting the median last 24h

But the overall median blocksize is still <100kb

So would be costly to miners to raise median limit

sech1: August 2017 iirc

the situation fluffypony described earlier was actually slighty worse - I only repeat for those that dont know some of the story

the PR with licensing issues was for concurrency speedups in the wallet, not ASM

the ASM speedups came later as a separate library for self-promotion. all of the comments from the original source were stripped, including original authors name

the ASM was converted from AT&T syntax to Intel syntax to further obscure the original source of the code

I remember all of that, he created a lot of buzz around that time and ended up starting ryo and integrating all that code there

so, aside from just grousing over the license, he was plagiarising the contribution?

I didn't know it was community funded, which makes it 100000 times worse

the changes made by him were primarily already done in my working branch that was public, but Im not sure if that was used as a reference

it took a while because I had to change some of the crypto code to make it work with dynamic libs and constant time

hyc yes, basically. the license was public domain so it was legal? but certainly unethical

vtnerd, did you ever bombard the wallet-rpc with tx requests to see when it would be saturated? (assuming you had all the outputs you ever needed)

sech1: actually, he joined SUMO, but later (According to him) found out that there was a hidden premine or something, and the community split into SUMO and Ryo

so much drama

Inge-, not eveRYOne wants to hear about that coin. :P

But hey, maybe someone got tired of the AHP's and moved on to try a spam attack instead

Mochi102 : I tested from daemon->wallet-cli not wallet-rpc->3rd party or were you asking about the former?

I've done various tests to determine limits of serialization code, but it was a bunch of direct timings. I have not tested to see full lmdb->wallet-cli timings

The PR has timings, like 17 minutes for 2 million blocks or so. That's something like 16 million outputs IIRC, very roughly.

my numbers on a 3900x were indicating that the crypto code may not be the bottleneck on bigger cache cpus. I didn't fully look into though

this was a 3900x, with a 970pro (but encrypted) and fast ram recommended for mining, so it was on the edge of what an "average" person would have

I was just wondering if it could handle 15-20 txs per second. Third party to wallet-rpc

what rpc call? one to retrieve received xmr, or ?

Spend

oh right that could have lots of results

Oh, sending ? No way.

1 tx/sec sounds more like it. It could be sped up if needed though.

oh wait I thought this was to retrieve "possible-spends" that was to send 15-20 txes a second?

*that was to send->, but this was for sending

yes spending xmr (transfer)

Spending's fairly inefficient. It constructs at least two txes to send one, for example.

yeah no clue on that, moo's guess is probably close

There are misc serialized daemon RPC calls to get data.

I see

yeah it probably hits the daemon-rpc to get the mapping for dummy outputs every time, which is a pretty response

*pretty big

id love a monero resources reporter, some data that shows what functions are using the most resources. i wonder if i could get it just by parsing a very verbose logfile

You can use perf and cachegrind.

sorry, thats ancillary. oh cool. ill check em out

"perf top -a" is very useful.

cause with ~150 wallets connected a node gets a bit sloggy

Dont connect 150 wallets :D

there's no limits for number of rpc connections

that's why I run monerod with ulimit -n

If you must offer public RPC, setup pay-for-rpc.

And see some asshole offer free rpc service so they can spy.

could this be related?

sethsimmons: gui is also tagged

Ah I’ll update for that then

Thanks

But no deterministic builds yet, correct?

CLI or GUI?

huh. 30% at fe_mul

GUI has no deterministic builds right?

they do exist now, dunno if the ones for 17.6 are ready

only on linux binaries, xiphon made a thread about it recently

The Linux GUI is reproducible, but we don’t have a repository for it like gitian.sigs.

sethsimmons: old.reddit.com/r/Monero/comments/k5mp0a

is it still using gitian build system?

GUI does not use gitian.

it is mostly a dockerfile

tobtoht: Do you or someone else want to post asking for sigs of that? I don’t know enough detail to really do it TBH

maybe better to keep GUI reproducible builds to IRC for now

Lyza: the peer id issue should be solved in v0.17.1.6

fantastic

26.6k transactions in the last 24h, it keeps growing...

Hello.. I was here before.. moneromooo

I still have trouble syncing..

What trouble?

Some people suggested the blockchain might be corrupted.. Are we certain about it?

sethsimmons my syncing instead of getting closer smaller, increases..

So im 2628 blocks behind and next i'm 2630 blocks behind..

What version are you running? What is the output of ‘status’ from your daemon?

did you delete the chain and sync again?

No.. What folder is the chain in?

sorry what OS?

sethsimmons Height: 2244517/2244517 (100.0%)

Thats strange..

+nioc im in ubuntu 20.04

sethsimmons: do you know where the chain is in linux?

~/.bitmonero/lmdb

2247151 is current height

Don’t need to delete I don’t think

You seem to be affected by the malicious peers

sethsimmons: moo looked at it previously, it is a non obvious issue

What version are you running?

Oh nvm then

Seems like he’s eclipsed

"sync_info" output would be interesting

nvm ? When i use status in monero-cli it says i'm synced..

it is not synced

You’re not, you’re behind the network in reality

can you run ./monerod, then in a second terminal enter ./monerod sync_info

and then post the output

Please provide the output of `sync_info` from your daemon

So why is status saying i'm synced?

OK +selsta

I’ll let selsta handle this :)

because your daemon does not know the real height so it thinks it is synced

Can I trust MyMonero app?

Thats the output of sync_info: pastebin.com/raw/kevL7a0Z

link it retarder

mymonero.com " We do not currently have an Android application. Please DO NOT download a MyMonero application from the Google Play store. Any MyMonero Android application is fake, and you will risk losing all your funds!"

I would try to sync from scatch

So what should I do..? Also what is nvm ?

Ok.. Where is the blockchain.. What folder files do i delete?

rm ~/.bitmonero/lmdb/data.mdb

Could i just delete the whole folder?

mrtrousers: yes, the failure to find output 1 of whatever amount it was means the db is borked.

Feel bork inc.® youtube.com/watch?v=muLAN-kP5pE

mrtrousers: yes

monero-wallet-cli tells you whether the wallet is synced with the daemon.

Will it be faster to download the blockchain raw?

Download yes, process it probably not.

Mmm, i dont understand..

But it's easy! youtube.com/watch?v=RXJKdh1KZ0w

You want to use a SSD if you can, and bump the number of out peers to, say, 40, from the defualt. It'll speed it up.

moneromooo should i just restart ./monerod then ? How do i bump up to 40?

Im using an SSD

"out_peers 40" in the monerod console

./monerod --out_peers 40 ?

Maybe. Check --help.

Well, it's starting from 0, cant believe it i spent all last night doing that.. I will try one more time..

I have used the --ban-list command too..

Hmm. Did you delete it already ?

moneromooo what means borked? Was that cause by the malicious nodes..?

It's odd if you got it yesterday and it's borked already...

No.

Yes i deleted it.. moneromoo

borked means broken.

Feel bork inc.® youtube.com/watch?v=muLAN-kP5pE

bork

borked

./monerod --out-peers 40 --ban-list /path/to/block.txt

would be the correct command ^

Ok, cools, thats what I did.. Will rebooting or restarting the computer while im downloading the blockchain and syncing break it?

If you end up with hte same problem, try: output_histogram @55704670000 0

It should output 2.

you wrote --out_peers that’s why I wrote it again

It should not, but if you use windows, it might.

Best to exit monerod before shutting down your machine, just in case.

force shutdown during sync can corrupt it

With Windows you're able to potentially break anything.

ah ok +selsta i wrote it right in the console, thanks.

Im in ubuntu

Could it be related to pruning the blockchain before being fully synced? moneromoo +selsta

how did you prune it?

./monero-blockchain-prune

moneromooo so would the command ./monerod output_histogram @55704670000 0

In theory, it should not... But it might be buggy.

have fixed the issue? Should i just restore the old blockchain and try it..?

If you want to prune, restart monerod now with --prnue-blockchain. It's *much* faster if you prune from the start (or near start).

No, it's a query only.

ok..

Is that command right: ./monerod --out-peers 40 --ban-list block.txt --prune-blockchain ?

Can't tell from the output if its pruning or not..

Yes, it's fine.

So pruning doesnt prune 1/3 of the blockchain, it prunes a 1/3 of each block.. right?

Lulz, i'm reading about pruning now.. anyways, thanks a lot for helping me with that issue, lulz

Pruning dumps signatures, which take... maybe 2/3 of a tx on average, *very* roughly. And it dumps that for 7/8ths of the chain,

All in all, it dumps about 2/3rds of the whole historical chain.

ie, 100 GB -> 30 GB.

Ok thanks for helping with the technical..

Is monero the coin with more presence at defcon?

I saw there is a monero village and stuff..

<agentpatience> lol wtf is a monero village

Yes there is a large Monero village at each DEFCON for the past few years

<xmrmatterbridge "<agentpatience> lol wtf is a mon"> Villages are the name for mini conferences at DEFCON each year with their own talks etc

0.17.1.6 now? xmrchain is a little trigger happy - binaries not yet on getmonero.org?

it is tagged, website release will be out tomorrow

Deterministic build verification is happening now so some early adopters are likely building from the tag and deploying.

cool

tl;dr or release notes out somewhere?

From reddit.com/r/Monero/comments/k8k94i…xygen_orion_has_been_tagged_you_may

Many more fixes of course but those are the high level changes

Lots of mitigations for the ongoing network attack.

also drop peers that decrease their target height and drop peers that don’t reply to queries

Nice 😀

and also drop peers that are idle for too long

hello

hi

nice work y'all!

so there was ways to automate it 😁

as i told u before this serialize_rctsig_base cant serialize block version > 3 but i cant replace with this one serialize_rctsig_prunable

when I left my node to sync

it had a segmentation fault 67% through

no other error, just "segmentation fault"

is there a way I can alert myself if this happens?

because I do not want this happening in production

Oh, that's very bad. Unless it's lmdb, which likes to SIGSEGV on bad data.

Try running in gdb.

Then, if it crashes again: bt

how do I run it in gdb?

"sudo gdb monerod --stagenet --data-dir /mypathhere" ?

I turned it back on already and it hasn't crashed yet

I'm somewhat aware that a segmentation fault is a programmer thing, not a user thing.

so I'm happy to test it

hmm gdb command not found

I have apt installing gdb

I don't normally use gdb but in this case there's good reason to do it

You don't need sudo .

THen:

try `gdb --args --stagenet --data-dir /mypathhere`

gdb monerod

run --stagenet --data-dir /mypathhere

Or what selsta said, yes.

ah wait, my comment was wrong :)

monerod is missing

`gdb --args monerod --stagenet --data-dir /mypathhere`

I do need sudo though because my directory needs root

the thing is I can't reproduce this bug

running that just brings me to the gdb prompt

run

ah ok cool

it's doing its thing now

if it crashes again enter "bt"

ok

it happened some time yesterday