| 00:08.02 | *** part/#bzflag op90 (~frans@sifar.xs4all.nl) |
| 00:53.02 | Constitution | TimRiker: any progress on the license? |
| 01:11.19 | Constitution | guess not |
| 01:28.21 | *** join/#bzflag Shuist (~Shuist@ppp203-122-213-220.static.internode.on.net) |
| 01:30.08 | blast007 | heh |
| 03:41.31 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 03:41.31 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 04:08.04 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 04:50.02 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 04:50.02 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 06:12.49 | *** join/#bzflag swigg (~Owner@97-82-254-128.dhcp.hckr.nc.charter.com) |
| 06:12.49 | *** join/#bzflag swigg (~Owner@bzflag/player/Swigg) |
| 06:58.42 | *** join/#bzflag swigg (~Owner@bzflag/player/Swigg) |
| 07:22.12 | *** join/#bzflag valera_rozuvan (~valera_ro@194.28.176.8) |
| 07:29.59 | *** join/#bzflag op90_ (~frans@sifar.xs4all.nl) |
| 08:01.45 | *** part/#bzflag op90_ (~frans@sifar.xs4all.nl) |
| 08:08.37 | *** join/#bzflag op90- (~frans@sifar.xs4all.nl) |
| 08:09.22 | *** join/#bzflag valera_rozuvan (~valera_ro@194.28.176.8) |
| 08:10.19 | *** join/#bzflag zuii (~akos@BC06BF93.catv.pool.telekom.hu) |
| 08:14.18 | *** part/#bzflag op90- (~frans@sifar.xs4all.nl) |
| 08:20.48 | *** join/#bzflag op90- (~frans@sifar.xs4all.nl) |
| 09:21.18 | *** join/#bzflag contempt (contempt@unaffiliated/contempt) |
| 09:40.57 | Notify | 03BZFlag:Op90 * 8944 /BZFlag_SVN: |
| 09:46.25 | Notify | 03BZFlag:Op90 * 8945 /BZFlag_Gi: Created page with "[[BZFlag_Git]] is used by the developers to keep track of changes to the sources of BZFlag. The sources for the most recent version of BZFlag can be found on [https://github...." |
| 09:50.56 | Notify | 03BZFlag:Op90 * 8946 /BZFlag_Gi: |
| 09:51.40 | Notify | 03BZFlag:Op90 * 8947 /BZFlag_Gi: |
| 10:02.56 | Notify | 03BZFlag:Op90 * 8948 /BZFlag_Gi: |
| 10:29.55 | *** join/#bzflag valera_rozuvan (~valera_ro@194.28.176.4) |
| 10:52.07 | *** join/#bzflag valera_rozuvan (~valera_ro@194.28.176.4) |
| 13:06.42 | *** join/#bzflag sanandres (~andres@181.49.254.102) |
| 13:57.30 | brad | ~bzgit |
| 13:57.30 | infobot | git clone --branch 2.4 https://github.com/BZFlag-Dev/bzflag.git |
| 14:18.44 | *** join/#bzflag sanandres (~andres@181.49.242.146) |
| 15:10.05 | *** join/#bzflag Flak (~Flak@2001:558:6030:27:cdbc:d363:dd9e:f5c3) |
| 15:12.23 | Flak | one question on bzflag |
| 15:13.09 | Flak | how is bzflag protected from DoS attacks |
| 15:14.14 | Flak | seems like that issue has not been addressed |
| 15:14.23 | op90 | Given the time it was first created my guess would be not. |
| 15:14.46 | brad | what kind of DoS? |
| 15:15.13 | brad | exploiting a bug in the software or flooding with traffic? |
| 15:15.26 | Flak | (D)enial (o)f (S)ervice attack |
| 15:16.03 | brad | there's more than one way to deny service |
| 15:16.38 | Flak | they are some tools out there |
| 15:17.16 | Flak | LAND, LOIC, HOIC, slowloris |
| 15:18.24 | brad | slowloris affects web servers only |
| 15:18.36 | Flak | i know |
| 15:18.44 | brad | loic would do nothing unless many people use it simultaneously |
| 15:18.46 | Flak | could be altered |
| 15:18.48 | brad | not sure what the others are |
| 15:20.01 | Flak | LAND (Local Area Network Denial) |
| 15:20.17 | Flak | The attack involves sending a spoofed TCP SYN packet (connection initiation) with the target host's IP address to an open port as both source and destination. |
| 15:20.34 | brad | if it only works on LAN, who cares? |
| 15:20.44 | brad | just punch the culprit ;) |
| 15:21.01 | Flak | the IP is spoofed |
| 15:21.14 | op90 | Does bzflag use UDP or TCP? |
| 15:21.20 | brad | both |
| 15:21.45 | Flak | see possible exploit |
| 15:21.53 | Flak | on the TCP |
| 15:22.13 | brad | afaik, you can't spoof tcp ip over the internet |
| 15:22.16 | brad | udp sure |
| 15:22.41 | brad | but bz uses tcp for auth (iirc) |
| 15:23.00 | brad | i'm sure blast will be able to give you more info when he's around |
| 15:23.05 | brad | or another dev |
| 15:23.06 | Flak | oh ok |
| 15:23.52 | op90 | Another attack vector would be the server that keeps track of all the game servers. |
| 15:24.04 | brad | yes |
| 15:24.25 | Flak | if one would hack that server they would be issues |
| 15:24.34 | brad | I believe that now has a (paid) DDoS protection service |
| 15:25.28 | Flak | ever heard of ReDoS |
| 15:25.41 | Flak | regular expression denial of service (ReDoS)[ |
| 15:27.46 | brad | that is relatively easy to fix though |
| 15:27.54 | brad | software fix |
| 15:29.08 | blast007 | "how is bzflag protected from DoS attacks" it isn't |
| 15:29.39 | Flak | one could use invaild instructions such as the pentium F00F bug |
| 15:29.56 | blast007 | if we're running it on an old Pentium .... |
| 15:30.01 | brad | well, it is in that some fixes have been put in place for the software bugs:) |
| 15:30.31 | brad | or do you only consider a DoS to be flooding with traffic |
| 15:30.49 | op90 | A more pertinent question might be how likely is bzflag to suffer from a DDoS attack. (Would imagine not very likely). |
| 15:30.59 | Flak | there are different kinds of DoS |
| 15:31.21 | blast007 | op90: it was attacked quite heavily for the last year or two, off and on |
| 15:31.45 | Flak | those would be concerns that i would have |
| 15:32.00 | op90 | Aha. that sucks |
| 15:32.21 | blast007 | <PROTECTED> |
| 15:32.54 | op90 | Disgrunteld players? |
| 15:33.07 | brad | bored kids |
| 15:33.11 | blast007 | one user in particular |
| 15:33.40 | blast007 | attacked the main server that I run, individual game servers, and individual users (including people's work connections) |
| 15:33.55 | op90 | Nasty! |
| 15:34.50 | blast007 | connection flooding (opening a lot of connections), actual exploits (crash bugs that would crash clients or servers, one that elevated them to a server admin), full on DDoS attacks through UDP reflection (chargen, DNS, NTP, SSDP, UPnP, etc) |
| 15:35.23 | op90 | From several addresses I presume? |
| 15:35.48 | blast007 | yeah, it was using other devices running those services to hit our systems |
| 15:36.02 | Flak | routers i guess |
| 15:36.27 | blast007 | spoof a UDP request packet to a service with a spoofed source IP, so the service responds to the spoofed IP (the target) instead, and with a much larger packet size |
| 15:37.50 | blast007 | they were actually renting botnet time at the end, so spending actual money to hit us |
| 15:38.02 | op90 | That is hard to protect against. |
| 15:38.15 | blast007 | and expensive |
| 15:38.24 | op90 | I can image |
| 15:38.35 | blast007 | we fixed a number of flaws in our code itself, but there's only so much we can do there |
| 15:40.20 | blast007 | I've got a better idea for how to design our services, but that's just an idea right now (it would involve one or more front line servers acting as load balancers, maybe with DDoS protection, and then hidden servers behind those) |
| 15:40.51 | Flak | the main issue is the with the User Datagram Protocol |
| 15:41.09 | Flak | the spoofing is easy with UDP |
| 15:41.33 | Flak | but that seems to be a smart idea |
| 15:41.49 | op90 | But it is needed for low latancy connections |
| 15:41.54 | blast007 | it would be harder if network owners configured stuff correctly |
| 15:42.16 | blast007 | http://spoofer.cmand.org/summary.php |
| 15:43.47 | op90 | and that is going to get worse with more and more things having an IP address... |
| 15:44.11 | blast007 | the Internet of horribly insecure Things |
| 15:44.24 | op90 | :) :) |
| 15:44.32 | blast007 | brought to you by cutrate software development! |
| 15:45.00 | Flak | agreed |
| 15:45.25 | Flak | i looked into one article |
| 15:46.17 | Flak | and it said that a refrigerator made spam emails |
| 15:48.24 | Flak | there also is metasploit framework |
| 15:48.40 | Flak | could also create issues |
| 15:48.42 | op90 | fridged probably never get updated |
| 15:48.50 | op90 | (friges) |
| 15:49.08 | Flak | they are way too easy to hack |
| 15:49.25 | blast007 | "Hold on, my fridge just tweeted me that I need to buy some milk" |
| 15:49.30 | Flak | lol |
| 15:50.17 | Flak | or oh no my fridge demands my email account password |
| 15:51.18 | blast007 | My fridge has encrypted all of my food and now I need to pay 0.1 bitcoins to get the decryption key. |
| 15:51.27 | Flak | lol |
| 15:51.39 | op90 | :) |
| 15:52.34 | op90 | the company I work for was almost hit by that. Guy it happened too could just in time jank the network cable |
| 15:52.48 | blast007 | heh |
| 15:52.49 | op90 | and prevent the network drives from being encrypted |
| 15:53.28 | blast007 | proper defense from that kind of attack is working backups, and backups done in such a way that your vulnerable systems don't have write access to the old backups |
| 15:53.54 | Flak | i used to work as a tech consultant |
| 15:54.03 | Flak | five years ago |
| 15:54.21 | blast007 | I block ads for everyone here to avoid a lot of the entry points for such an attack |
| 15:54.38 | blast007 | and nobody has admin rights on their computers except network staff |
| 15:55.04 | op90 | agree, and we do have backups but it still costs time to roll back the back up. meanwhile people sitting idle. |
| 15:55.06 | Flak | almost all the servers were taken down then. |
| 15:55.35 | Flak | the main issue with backups is possible theft |
| 15:55.49 | Flak | from an insider |
| 15:56.36 | blast007 | physical security is important too :) |
| 15:56.44 | Flak | agreed |
| 15:57.06 | op90 | or while they are on there way to the remote location |
| 15:58.19 | Flak | once had to use the servers to make a DDoS attack toward some hackers that were trying to acess passwords. |
| 15:58.52 | blast007 | retalitory attacks are bad and just as illegal |
| 15:59.01 | op90 | douns like you had a fun job :) |
| 15:59.06 | op90 | (sounds) |
| 15:59.24 | blast007 | most people trying to break in don't use their own systems either, so you attacked some innocent (though inscure) bystander |
| 15:59.42 | brad | blast007: even in US? aren't you allowed to shoot people who come into your property? ;D |
| 16:00.40 | Flak | it was so close to compromise our mainframes |
| 16:02.33 | Flak | losses would have caused bankruptcy |
| 16:03.47 | Flak | of course the servers were offshore |
| 16:05.24 | Flak | so US laws did not apply to our case |
| 16:05.58 | blast007 | https://www.youtube.com/watch?v=s1cxNYNkJWs |
| 16:10.05 | Flak | blast007: lol nice vid |
| 16:10.08 | op90 | :) |
| 16:10.48 | Flak | i also checked with my lawyer just to be sure |
| 16:13.08 | op90 | Got to go dinner is waiting. Till later. |
| 16:13.14 | blast007 | cya |
| 16:13.21 | Flak | cu |
| 16:13.26 | *** part/#bzflag op90 (~frans@sifar.xs4all.nl) |
| 17:16.21 | *** join/#bzflag zuii (~akos@BC06BF93.catv.pool.telekom.hu) |
| 17:46.21 | *** join/#bzflag sanandresm (~andres@181.48.26.211) |
| 17:55.26 | *** join/#bzflag alpha1-2 (~alpha1@host30.190-138-236.telecom.net.ar) |
| 18:00.01 | *** join/#bzflag contempt (contempt@unaffiliated/contempt) |
| 18:30.53 | *** join/#bzflag op90 (~frans@sifar.xs4all.nl) |
| 18:35.06 | allejo | ~alezakos++ |
| 18:35.13 | allejo | flagOnSpawn <3 |
| 18:41.15 | *** join/#bzflag DTRemenak|RDP (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 18:41.15 | *** mode/#bzflag [+v DTRemenak|RDP] by ChanServ |
| 18:46.17 | alezakos | oh |
| 18:46.30 | alezakos | that's veeeeery old :P |
| 18:46.54 | allejo | it is but it works :p |
| 18:47.48 | alpha1-2 | allejo: read GU :P |
| 19:16.57 | *** join/#bzflag JeffM (~Jeffrey@2605:e000:7d02:c700:addd:cd34:8512:2507) |
| 19:17.09 | JeffM | you want me to test what on AMD now? |
| 19:17.26 | allejo | hai jeff! |
| 19:17.48 | JeffM | I forgot I banned my irccloud account :) |
| 19:18.35 | JeffM | your RC2 runs fine with a simple test on widnows 10 with a radeon HD 7950 |
| 19:18.45 | JeffM | connected to a server, did stuff, quit |
| 19:20.10 | JeffM | it was running full screen 1440x900 |
| 19:21.46 | blast007 | hrm, then you're not getting the issue ;) this was failing before even the main menu showed up |
| 19:22.07 | blast007 | could still see through the window, in fact |
| 19:22.08 | JeffM | does resolution have anything to do with it? |
| 19:22.19 | blast007 | sometimes |
| 19:22.41 | JeffM | I can hook it up to a 1080p display |
| 19:22.43 | blast007 | I've had it more reliably broken with a 1920x1080 screen |
| 19:22.46 | blast007 | most* |
| 19:23.21 | blast007 | you might have to log off/on or reboot after the display has been attached and detected to get it to happen |
| 19:23.35 | *** join/#bzflag DTRemenak (~DTRemenak@cpe-172-112-28-185.socal.res.rr.com) |
| 19:23.35 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 19:23.35 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 19:23.46 | blast007 | if you change scaling or resolution while in a session, it will run fine - or if attach a screen after logged it |
| 19:24.33 | JeffM | I'll attach and reboot |
| 19:24.43 | blast007 | k |
| 19:25.43 | *** join/#bzflag sanandresm (~andres@181.49.254.102) |
| 19:26.30 | JeffM | guhhh, AMD installs so much crapware |
| 19:27.25 | blast007 | yeah... I left off the 'gaming evolved' thing |
| 19:28.12 | JeffM | you want standard old pre-darwinian gaming? |
| 19:29.06 | JeffM | yeah, runs fine |
| 19:29.12 | blast007 | bleh :P |
| 19:29.41 | JeffM | describe the problem to me |
| 19:30.57 | blast007 | I run the game, the icon shows up at the bottom, cursor goes into the spinner, but I still see everything else on my screen |
| 19:31.18 | blast007 | click once, goes black. Click again, goes gray (and it says the program isn't responding, do you want to close). |
| 19:31.29 | JeffM | so it can't create the context |
| 19:31.40 | blast007 | It's getting stuck inside DestroyWindow when it's killing the child window. |
| 19:32.09 | JeffM | why is it destroying a window on startup? |
| 19:32.40 | JeffM | he asks about the code he wrote in 1996 |
| 19:32.44 | blast007 | :) |
| 19:33.04 | JeffM | do you have a debugger on the system with the problem? |
| 19:33.08 | blast007 | yes |
| 19:33.11 | blast007 | 2010 |
| 19:33.30 | JeffM | put a breakpoint before destory window and see what in the callstack is getting there |
| 19:33.58 | JeffM | there is a cheezy thing you can do, throw a thread.sleep and a window message pump between destorying and doing anything else |
| 19:34.04 | JeffM | to let the driver "do stuff" |
| 19:38.38 | blast007 | that system is at home, but I think it was destroying the child because it was going fullscreen (think it starts out with a 640x480 parent window), so WM_DISPLAYCHANGE gets called, which ends up calling WinWindow::deactivateAll(), which calls WinWindow::freeContext(), which calls WinWindow::destroyChild(), which calls DestroyWindow(hwndChild) (which it never comes back from) |
| 19:39.17 | JeffM | what does that windows winproc do on the close event? |
| 19:39.33 | JeffM | effin winprocs.... |
| 19:40.04 | blast007 | https://github.com/BZFlag-Dev/bzflag/files/72570/trace01.txt does that help? |
| 19:40.50 | blast007 | I don't get another message after WM_NCDESTROY |
| 19:41.09 | JeffM | well you don't process another message after that |
| 19:41.17 | JeffM | so the question is what does the code do when it gets that |
| 19:41.36 | JeffM | that's all in platform right? |
| 19:44.58 | JeffM | so in WinDisplay.cxx you can put a breakpoint in LONG WINAPIWinDisplay::Rep::windowProc(HWND hwnd, UINT msg....)in the WM_DESTORY case and see if it gets in there. |
| 19:45.06 | JeffM | if it doens't then it ain't your code that dies |
| 19:45.21 | JeffM | if it does, then you can step into the window class and see where it poops |
| 19:45.40 | JeffM | you can probalby get away with out the child window |
| 19:45.44 | JeffM | but that's a larger change. |
| 19:46.49 | JeffM | it's also possible that WM_ACTIVATE may be getting in the way, it's possible you are sending an deactivate event before or after close |
| 19:47.09 | JeffM | the code can ignore activation events if it knows the window in the process of being destroyed |
| 19:47.59 | JeffM | the simpelset way to do that would be to set BZDB.isTrue("Win32NoMin") to true before the destory, then resore it after. |
| 19:50.17 | *** join/#bzflag sanandresm (~andres@181.48.26.211) |
| 20:26.44 | JeffM | well, I'll have the card for a few days, if you want me to try anything else, shoot me a line |
| 20:30.49 | alpha1-2 | blast007: filter help page 1: "this filter would find any 2 or 3 shot servers that have at least one player online: /p>1,s>1,s<4 (...)". It should say '/p>0' or '/p>=1'. It's a common mistake :) |
| 20:36.52 | the_map | I note that in the next paragraph it notes that "player count must be greater than 1" about that filter |
| 20:40.34 | alpha1-2 | thanks, I have not checked all, just wanted to give that one I just saw |
| 21:28.20 | *** part/#bzflag Flak (~Flak@2001:558:6030:27:cdbc:d363:dd9e:f5c3) |
| 22:55.51 | *** join/#bzflag alpha1-2 (~alpha1@host30.190-138-236.telecom.net.ar) |
| 23:05.37 | *** part/#bzflag op90 (~frans@sifar.xs4all.nl) |
| 23:28.58 | *** join/#bzflag meeba (~lamer@c-73-14-146-7.hsd1.co.comcast.net) |