The Official Status Thread


  • Notification Spam Recipient

    In other news, literally every single UE4 VR title I launched through Steam is 100 percent incapable of showing the steam overlays. Fuck.



  • STATUS

    Spent hours yesterday writing my own heatmap generating algorithm. Finally done. It takes it 15 seconds to render an image.

    Look into the source code of an OSS heatmap implementation.

    "Oh, so they diffuse the data using native canvas radial blur, so it gets done in a shader instead of RAM".

    Modify my thing to do it like that. Generates heatmap in 2-3 seconds.

    The moral: I should NOT write algorithms.



  • @rhywden said in The Official Status Thread:

    Small addendum: I found out that if you accrue a number of parking tickets the responsible governmental agency may just decide to call into question your ability to drive. They may then order you to undergo a mandatory psychological check (called the MPU which is colloquially known as the Morons' Test) which is not easy (in fact. 50% of attendees fail it).
    And if you fail that test your driving license is gone. At least until you manage to pass the test.

    Story for https://www.reddit.com/r/ProRevenge/ ?



  • @polygeekery said in The Official Status Thread:

    Status: So this just happened:

    πŸ‘© "Oh yeah, we are supposed to have a solar eclipse soon."
    polygeekery "Yeah."
    πŸ‘© "It's during the day though, isn't it?"
    polygeekery "..."
    πŸ‘© "What?"
    polygeekery "Just think about what you just said."
    πŸ‘© "What?"
    polygeekery "It wouldn't be much of a solar eclipse for us if it happened at night."

    Really, Polygeekery. How long have you been married? Long enough to know that one Does. Not. Use. Logic. On. Ones. Wife!



  • @ben_lubar Here, an "arbitrary number" is always 42. An "arbitrary large number" is always 4711.



  • @cartman82 said in The Official Status Thread:

    @rhywden said in The Official Status Thread:

    Small addendum: I found out that if you accrue a number of parking tickets the responsible governmental agency may just decide to call into question your ability to drive. They may then order you to undergo a mandatory psychological check (called the MPU which is colloquially known as the Morons' Test) which is not easy (in fact. 50% of attendees fail it).
    And if you fail that test your driving license is gone. At least until you manage to pass the test.

    Story for https://www.reddit.com/r/ProRevenge/ ?

    Not yet. It's more of a long game.


  • 🚽 Regular

    @ben_lubar said in The Official Status Thread:

    @cursorkeys said in The Official Status Thread:

    Status: Adding language support and I thought of @ben_lubar

    else if (langId.equals("jbo"))
            {
                return LOJBAN;
            }
    

    Machine-translate only for that one I'm afraid but it's going in. For testing for formatting issues obviously, not because I'm having fun.

    Why not use a HashMap or a TreeMap?

    Hmm, good idea, thanks. I'll stick a //TODO: on it for refactoring. I guess I have a bit of an aversion to the map collections.


  • Java Dev

    @polygeekery said in The Official Status Thread:

    Status: So this just happened:

    πŸ‘© "Oh yeah, we are supposed to have a solar eclipse soon."
    polygeekery "Yeah."
    πŸ‘© "It's during the day though, isn't it?"
    polygeekery "..."
    πŸ‘© "What?"
    polygeekery "Just think about what you just said."
    πŸ‘© "What?"
    polygeekery "It wouldn't be much of a solar eclipse for us if it happened at night."

    Can we get it moved so it happens during our daytime instead?


  • 🚽 Regular

    Status: :wtf: I'm at work so I'm not clicking it to find out.

    0_1502969757805_bc363176-7414-4d2c-b9c3-eae00a2439fc-image.png



  • @polygeekery
    It could be in the evening, which is distinct from the day in that most normal people are off work & school.


  • Discourse touched me in a no-no place

    @cursorkeys said in The Official Status Thread:

    @ben_lubar said in The Official Status Thread:

    @cursorkeys said in The Official Status Thread:

    Status: Adding language support and I thought of @ben_lubar

    else if (langId.equals("jbo"))
            {
                return LOJBAN;
            }
    

    Machine-translate only for that one I'm afraid but it's going in. For testing for formatting issues obviously, not because I'm having fun.

    Why not use a HashMap or a TreeMap?

    Hmm, good idea, thanks. I'll stick a //TODO: on it for refactoring. I guess I have a bit of an aversion to the map collections.

    You can also switch on strings with recent-enough Java. That's (probably) going to do something efficient internally.




  • Java Dev

    STATUS: Rough year planning done! I can celebrate with going home from work on time! πŸ‘―β™‚


  • BINNED

    Status: I hate localizing software. Especially to my own language:

    0_1502979285808_c54d9695-f1c9-44be-9876-e94d87694b8b-image.png

    Ok, wrong grammatical case there, Google, but yeah, that's it.

    0_1502979351233_4c82f382-5758-4dbb-aa7e-6ae06fa3161a-image.png

    Yup.

    To be clear, I went to ask Google just to make sure I'm not completely illiterate. There is no word I can use there. Awesome.


  • FoxDev

    @onyx said in The Official Status Thread:

    Status: I hate localizing software. Especially to my own language:

    0_1502979285808_c54d9695-f1c9-44be-9876-e94d87694b8b-image.png

    Ok, wrong grammatical case there, Google, but yeah, that's it.

    0_1502979351233_4c82f382-5758-4dbb-aa7e-6ae06fa3161a-image.png

    Yup.

    To be clear, I went to ask Google just to make sure I'm not completely illiterate. There is no word I can use there. Awesome.

    It's a Lexical Lacuna!


  • BINNED

    @accalia said in The Official Status Thread:

    It's a Lexical Lacuna!

    I'm considering making E_LANGUAGE_INSUFFICIENT a translation for callee.


  • Java Dev

    @cartman82 said in The Official Status Thread:

    STATUS

    Spent hours yesterday writing my own heatmap generating algorithm. Finally done. It takes it 15 seconds to render an image.

    Look into the source code of an OSS heatmap implementation.

    "Oh, so they diffuse the data using native canvas radial blur, so it gets done in a shader instead of RAM".

    Modify my thing to do it like that. Generates heatmap in 2-3 seconds.

    The moral: I should NOT write algorithms.

    I'm always hesitant to do that for fear of getting bit by the GPL vampire.


  • kills Dumbledore

    @onyx No decent translations from alternative phrases like "Call Recipient"?


  • BINNED

    @jaloopa That could work I guess, but considering this is to be shown in a table and/or in an exported/printed document translating something short-ish Callee name to Ime primatelja poziva seems unsightly to me πŸ€·β™‚

    Dunno, you might argue that the name might be long anyway, but... ugh. May be my only way out of this mess though.


  • FoxDev

    @onyx said in The Official Status Thread:

    @accalia said in The Official Status Thread:

    It's a Lexical Lacuna!

    I'm considering making E_LANGUAGE_INSUFFICIENT a translation for callee.

    Lexical Lacuna is more poetic than throw new LanguageInsufficientException();


  • BINNED

    @accalia said in The Official Status Thread:

    Lexical Lacuna is more poetic

    It fails to express my rage though. Error in all caps at least somewhat helps. :p


  • FoxDev

    maybe this then?

    throw new LanguageInsufficientException("FUCK YOU LANGUAGE! GIVE ME WORDS I CAN USE!");



  • @onyx said in The Official Status Thread:

    Yup.
    To be clear, I went to ask Google just to make sure I'm not completely illiterate. There is no word I can use there. Awesome.

    Nothing I can think of in Serbian either.

    Pozivoprimac? :)



  • @onyx just randomly write one string in romanized hebrew. It worked for PHP!



  • @rhywden said in The Official Status Thread:

    They may then order you to undergo a mandatory psychological check (called the MPU which is colloquially known as the Morons' Test) which is not easy (in fact. 50% of attendees fail it).

    Presumably only suspected morons have to do the test, though? If it's meant to be hard for morons, that sounds like it works as intended.



  • Status: Apparently not escaping user input managed to bring our whole site down. Also, we're doing all of our security work now, after the first release to customers. Everything is wonderful and none of this was designed by incompetent morons.



  • @magus said in The Official Status Thread:

    @rhywden said in The Official Status Thread:

    They may then order you to undergo a mandatory psychological check (called the MPU which is colloquially known as the Morons' Test) which is not easy (in fact. 50% of attendees fail it).

    Presumably only suspected morons have to do the test, though? If it's meant to be hard for morons, that sounds like it works as intended.

    Yes, people who were caught speeding, DUI or other incidents with a non-negligible fine will usually have to face this test (better word would be: psychological exam - it's not a simple questionnaire, after all) quite soon. I.e., indeed persons who behave in a way that casts doubts on their ability to participate in street traffic in a harmless manner.



  • @magus Addendum: From what I understand, someone could inject arbitrary code into our production server.

    They say the team I'm on will fix it, since we're the ones adding security to the system.


  • Notification Spam Recipient

    @magus said in The Official Status Thread:

    @magus Addendum: From what I understand, someone could inject arbitrary code into our production server.

    They say the team I'm on will fix it, since we're the ones adding security to the system.

    Perfectly cromulent, logic is sound!


  • :belt_onion:

    @cartman82 said in The Official Status Thread:

    The moral: I should NOT write algorithms.

    This is true for most people, TBH. Writing good algorithms requires a ton of mathematical knowledge, generally, and a fuckton of research.


  • 🚽 Regular

    @dkf said in The Official Status Thread:

    @cursorkeys said in The Official Status Thread:

    @ben_lubar said in The Official Status Thread:

    @cursorkeys said in The Official Status Thread:

    Status: Adding language support and I thought of @ben_lubar

    else if (langId.equals("jbo"))
            {
                return LOJBAN;
            }
    

    Machine-translate only for that one I'm afraid but it's going in. For testing for formatting issues obviously, not because I'm having fun.

    Why not use a HashMap or a TreeMap?

    Hmm, good idea, thanks. I'll stick a //TODO: on it for refactoring. I guess I have a bit of an aversion to the map collections.

    You can also switch on strings with recent-enough Java. That's (probably) going to do something efficient internally.

    The SMALI is interesting for a simple two-case switch on a string:

    public class test
    {
    	Object theThing = new Object();
    
    	Object test()
    	{
    		String testString = "test";
    	
    		switch (testString)
    		{
    			case "nottest":
    				return null;
    			case "test":
    				return theThing;
    		}
    		
    		return null;
    	}
    }
    
    .class public Ltest;
    .super Ljava/lang/Object;
    .source "test.java"
    
    
    # instance fields
    .field theThing:Ljava/lang/Object;
    
    
    # direct methods
    .method public constructor <init>()V
        .registers 2
    
        .prologue
        .line 1
        invoke-direct {p0}, Ljava/lang/Object;-><init>()V
    
        .line 3
        new-instance v0, Ljava/lang/Object;
    
        invoke-direct {v0}, Ljava/lang/Object;-><init>()V
    
        iput-object v0, p0, Ltest;->theThing:Ljava/lang/Object;
    
        return-void
    .end method
    
    
    # virtual methods
    .method test()Ljava/lang/Object;
        .registers 5
    
        .prologue
        const/4 v0, 0x0
    
        .line 7
        const-string v2, "test"
    
        .line 9
        const/4 v1, -0x1
    
        invoke-virtual {v2}, Ljava/lang/String;->hashCode()I
    
        move-result v3
    
        sparse-switch v3, :sswitch_data_26
    
        :cond_b
        :goto_b
        packed-switch v1, :pswitch_data_30
    
        .line 17
        :goto_e
        :pswitch_e
        return-object v0
    
        .line 9
        :sswitch_f
        const-string v3, "nottest"
    
        invoke-virtual {v2, v3}, Ljava/lang/String;->equals(Ljava/lang/Object;)Z
    
        move-result v2
    
        if-eqz v2, :cond_b
    
        const/4 v1, 0x0
    
        goto :goto_b
    
        :sswitch_19
        const-string v3, "test"
    
        invoke-virtual {v2, v3}, Ljava/lang/String;->equals(Ljava/lang/Object;)Z
    
        move-result v2
    
        if-eqz v2, :cond_b
    
        const/4 v1, 0x1
    
        goto :goto_b
    
        .line 14
        :pswitch_23
        iget-object v0, p0, Ltest;->theThing:Ljava/lang/Object;
    
        goto :goto_e
    
        .line 9
        :sswitch_data_26
        .sparse-switch
            0x364492 -> :sswitch_19
            0x7ef04de5 -> :sswitch_f
        .end sparse-switch
    
        :pswitch_data_30
        .packed-switch 0x0
            :pswitch_e
            :pswitch_23
        .end packed-switch
    .end method
    
    

    If you add more cases it just adds another line to each sparse-switch and packed-switch and more String.equals tests. Same result if you make each case return a separate object.
    I was expecting some optimisation. It was all going so well to begin with, a compare by string hash then it goes all potato.


  • @tsaukpaetra said in The Official Status Thread:

    so I'm stuck with .ToList()ing the db.CHAR_Parts section, completely obviating the point of sticking these things into the one table!

    FML :angry:

    Sorry if this is coming way too late to be useful, but I think there's something in the Fluent API about configuring different classes to use the same table for storage, provided you have a discriminator column, such as PartType. That probably won't help with having each type of part live in its own property in the model, but you might be able to replace those properties with computed ones that .OfType<>() on the query or its reification.

    @raceprouk said in The Official Status Thread:

    And their 2D efforts have been lamentable too: Sonic 4 is an embarrassment.

    Dimps, not Sonic Team. A company widely known for their buggy physics engines and yet Sega keeps picking them anyway. Subcontracting reigns eternal...


  • Notification Spam Recipient

    @twelvebaud said in The Official Status Thread:

    @tsaukpaetra said in The Official Status Thread:

    so I'm stuck with .ToList()ing the db.CHAR_Parts section, completely obviating the point of sticking these things into the one table!

    FML :angry:

    Sorry if this is coming way too late to be useful, but I think there's something in the Fluent API about configuring different classes to use the same table for storage, provided you have a discriminator column, such as PartType. That probably won't help with having each type of part live in its own property in the model, but you might be able to replace those properties with computed ones that .OfType<>() on the query or its reification.

    Not too late, at the moment it's still only one class.

    What is this "fluent API" you speak of, and how well does it interact with Entity Framework 5?


  • FoxDev

    @twelvebaud said in The Official Status Thread:

    @raceprouk said in The Official Status Thread:

    And their 2D efforts have been lamentable too: Sonic 4 is an embarrassment.

    Dimps, not Sonic Team. A company widely known for their buggy physics engines and yet Sega keeps picking them anyway. Subcontracting reigns eternal...

    Ah yes.

    Hard to believe it's the same Dimps who did a good job with the Advance trilogy and the Rush games.



  • @tsaukpaetra In your DbContext, you can override the OnModelCreating(ModelBuilder) function to change and annotate the model EF thought you would use. It looks like what you're trying to do is most similar to moving from table-per-type to table-per-hierarchy, which I think works with EF5.


  • Notification Spam Recipient

    Status: WTF, Hard Drive, why are you going so slow?!?

    0_1502991326562_0cd6ec65-b253-4a59-803f-3cac73d833d2-image.png

    It's the windows 10 ISO file I'm loading into a Ramdisk!

    Wait...

    0_1502991476434_b54f2c4e-b105-4ab0-be1d-709d544cfba7-image.png

    WTFTDEMSYR Imma slap a disk here in a minute...


  • Notification Spam Recipient

    @tsaukpaetra said in The Official Status Thread:

    Status: Wondering how I'm going to troubleshoot an image that boots and runs fine in Hyper-V and VirtualBox, but doesn't in QEMU (what the OnApp server is apparently using for its virtualization)...

    Ok, using the QEMU that the WinBuilder package I'm using came with (I guess 0.9.0?) I get lovely bugchecks like so:

    
    Connected to Windows 10 14393 x64 target at (Thu Aug 17 13:27:12.271 2017 (UTC - 7:00)), ptr64 TRUE
    Kernel Debugger connection established.
    Symbol search path is: srv*
    Executable search path is: 
    Windows 10 Kernel Version 14393 MP (1 procs) Free x64
    Built by: 14393.0.amd64fre.rs1_release.160715-1616
    Machine Name:
    Kernel base = 0xfffff800`6468d000 PsLoadedModuleList = 0xfffff800`64992060
    System Uptime: 0 days 0:00:00.022
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    KDTARGET: Refreshing KD connection
    
    *** Fatal System Error: 0x0000005d
                           (0x00000000078BFBFD,0x000000002191ABFD,0x0000000000000000,0x0000000000000000)
    
    Break instruction exception - code 80000003 (first chance)
    
    A fatal system error has occurred.
    Debugger entered on first try; Bugcheck callbacks have not been invoked.
    
    A fatal system error has occurred.
    
    Connected to Windows 10 14393 x64 target at (Thu Aug 17 13:27:14.349 2017 (UTC - 7:00)), ptr64 TRUE
    Loading Kernel Symbols
    ...............................................................
    .......
    
    Press ctrl-c (cdb, kd, ntsd) or ctrl-break (windbg) to abort symbol loads that take too long.
    Run !sym noisy before .reload to track down problems loading symbols.
    
    .........................................................
    .
    Loading User Symbols
    
    Cannot read PEB32 from WOW64 TEB32 f000ff53 - NTSTATUS 0xC0000141
    *******************************************************************************
    *                                                                             *
    *                        Bugcheck Analysis                                    *
    *                                                                             *
    *******************************************************************************
    
    Use !analyze -v to get detailed debugging information.
    
    BugCheck 5D, {78bfbfd, 2191abfd, 0, 0}
    
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    Probably caused by : ntkrnlmp.exe ( nt! ?? ::OKHAJAOM::`string'+38c1 )
    
    Followup:     MachineOwner
    ---------
    
    nt!DbgBreakPointWithStatus:
    fffff800`647dca60 cc              int     3
    kd> g
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    GetContextState failed, 0xD0000001
    Break instruction exception - code 80000003 (first chance)
    
    A fatal system error has occurred.
    Debugger entered on first try; Bugcheck callbacks have not been invoked.
    
    A fatal system error has occurred.
    
    nt!DbgBreakPointWithStatus:
    fffff800`647dca60 cc              int     3
    
    
    

    Some magical googling suggests

    So I said "Fuck it!" and checked if there was a newer version of QEmu.

    There is, and it seems to be one that's not a decade old, so trying that:

    Microsoft (R) Windows Debugger Version 10.0.15063.468 AMD64
    Copyright (c) Microsoft Corporation. All rights reserved.
    
    Opened \\.\com1
    Waiting to reconnect...
    Connected to Windows 10 14393 x64 target at (Thu Aug 17 13:20:17.165 2017 (UTC - 7:00)), ptr64 TRUE
    Kernel Debugger connection established.
    Symbol search path is: srv*
    Executable search path is: 
    Windows 10 Kernel Version 14393 MP (1 procs) Free x64
    Built by: 14393.0.amd64fre.rs1_release.160715-1616
    Machine Name:
    Kernel base = 0xfffff800`62a7e000 PsLoadedModuleList = 0xfffff800`62d83060
    System Uptime: 0 days 0:00:00.022
    IOINIT: Built-in driver \Driver\sacdrv failed to initialize with status - 0xC0000037
    We are running at normal mode.
    refs.sys loaded.
    refs.sys loaded.
    KDTARGET: Refreshing KD connection
    ******************************************************************
    * This break indicates this binary is not signed correctly: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\syswow64\comdlg32.dll
    * and does not meet the system policy.
    * The binary was attempted to be loaded in the process: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\System32\smss.exe
    * This is not a failure in CI, but a problem with the failing binary.
    * Please contact the binary owner for getting the binary correctly signed.
    ******************************************************************
    Break instruction exception - code 80000003 (first chance)
    CI!CipReportAndReprieveUMCIFailure+0x45f:
    fffff80c`295ff367 cc              int     3
    kd> g
    ******************************************************************
    * This break indicates this binary is not signed correctly: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\syswow64\lpk.dll
    * and does not meet the system policy.
    * The binary was attempted to be loaded in the process: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\System32\smss.exe
    * This is not a failure in CI, but a problem with the failing binary.
    * Please contact the binary owner for getting the binary correctly signed.
    ******************************************************************
    Break instruction exception - code 80000003 (first chance)
    CI!CipReportAndReprieveUMCIFailure+0x45f:
    fffff80c`295ff367 cc              int     3
    kd> g
    ******************************************************************
    * This break indicates this binary is not signed correctly: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\syswow64\normaliz.dll
    * and does not meet the system policy.
    * The binary was attempted to be loaded in the process: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\System32\smss.exe
    * This is not a failure in CI, but a problem with the failing binary.
    * Please contact the binary owner for getting the binary correctly signed.
    ******************************************************************
    Break instruction exception - code 80000003 (first chance)
    CI!CipReportAndReprieveUMCIFailure+0x45f:
    fffff80c`295ff367 cc              int     3
    kd> g
    ******************************************************************
    * This break indicates this binary is not signed correctly: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\syswow64\wldap32.dll
    * and does not meet the system policy.
    * The binary was attempted to be loaded in the process: \Device\Ramdisk{d9b257fc-684e-4dcb-ab79-03cfa2f6b750}\Windows\System32\smss.exe
    * This is not a failure in CI, but a problem with the failing binary.
    * Please contact the binary owner for getting the binary correctly signed.
    ******************************************************************
    Break instruction exception - code 80000003 (first chance)
    CI!CipReportAndReprieveUMCIFailure+0x45f:
    fffff80c`295ff367 cc              int     3
    kd> g
    

    Stuff's not signed, woohoo, but hey, it boots!

    Time to write an inquiry...



  • @cartman82 said in The Official Status Thread:

    STATUS

    Spent hours yesterday writing my own heatmap generating algorithm. Finally done. It takes it 15 seconds to render an image.

    Look into the source code of an OSS heatmap implementation.

    "Oh, so they diffuse the data using native canvas radial blur, so it gets done in a shader instead of RAM".

    Modify my thing to do it like that. Generates heatmap in 2-3 seconds.

    The moral: I should NOT write algorithms.

    This reminds me of when I was doing undergraduate research. I had some atomic force microscope data (a height map of a rough glass surface) and I needed to find the size and period of the bumps. So, I write a 2D fourier transform algorithm basically straight from the textbook definition (probably O(n^4)). For the 256x256 dataset, the calculation took 8 hours. When I presented the results the next day and mentioned how long it took, one of the other researches asked, "Haven't you heard of an FFT?"

    After including FFTW, my program took 2 seconds. A 15,000x speedup is pretty good, right?


  • Notification Spam Recipient

    Status: OMFGWTFBBQ this is supposed to be a tool I use to help figure out why applications aren't loading!

    0_1503019880523_e863abe2-ec22-4bef-bd35-c7f39bf55d04-image.png

    Gah!


  • FoxDev

    @tsaukpaetra At least you can start Procmo- Wait…


  • Notification Spam Recipient

    @tsaukpaetra said in The Official Status Thread:

    slap a disk

    Main commonly-accessed folders slapped, now working on the less-frequently-accessed files...

    0_1503020500146_5d17c2b9-9732-462a-a6c2-844b885a52a4-image.png


  • Notification Spam Recipient

    @raceprouk said in The Official Status Thread:

    @tsaukpaetra At least you can start Procmo- Wait…

    And on a machine I've profiled the target app on where said app is actually working correctly:

    0_1503020663998_9f0bdd5e-ae50-4697-bfc5-d1e75d6875e5-image.png

    ....

    I killed the app forever ago, why is it eating RAM?

    Telling it to stop capturing events (whoops) immediately let it slide back down to ~37 Mb.


  • Notification Spam Recipient

    @tsaukpaetra said in The Official Status Thread:

    why is it

    Getting closer!

    0_1503022111346_27a68c1a-fee4-4fed-b5cc-d90ef0b7ce63-image.png


  • Grade A Premium Asshole

    Status: Was RDP'd in to a client location checking the logs trying to track down an issue for them. There is a lot of noise in the logs referring to an ethernet interface that we have not used for quite a while so I decide to just delete the interface. I give it a bit of thought, verified that it is no longer in usage, thought some more to consider any way this could impact them and release the role.

    RDP drops. All of their machines start dropping out of our remote support system. I cannot get connected in.

    http://replygif.net/i/405.gif

    I text the owner, he verifies that their internet is completely down.

    Fuckity fuck fuck fuck. I run upstairs, start to change (My morning started with someone calling at 8am and has not stopped yet, so I never got dressed) and finally my mind gets extricated from Windows land. Back downstairs to try the firewall on SSH. Connected. Set the interface back to what it was, reboot the firewall for good measure and everything comes back online.


  • Banned

    This post is deleted!

  • β™Ώ (Parody)

    Motherfucker.


  • Java Dev

    STATUS: Ending the work week with throwing a couple Windows Server 2008 DVDs in the waste bin. Also, as I ended up the last person here I'm also stuck waiting for the carpenters to finish in the electronics lab so I can lock up the place after they leave.

    I have also perused more documentation during the day and found even more evidence of the author of said documentation being completely clueless. The sheer number of WTF and outdated information, as well as plenty inconsistencies in the material has left several palm-sized indentations in my face and potentially fractured my forehead. I wish I could just take one (or all) of the broken laptops on my desk and just throw them out the window to calm down.

    The server documentation is based off Windows Server 2008 or Windows Server 2012 depending on which one you read. Which is extra useful to me, as I don't have any Windows servers to manage here, only Linux servers. There are also installation and configuration instructions for Windows 7 and Linux Ubuntu 10. With an added promise of an update for Linux Ubuntu 12 when that is released.

    Oh $deity. I just realized. I thought the list of suggested use areas for the Intel CPU lineup was a bit weird as it did the Core i7/i5/i3 and then Pentium 4. But it's the author thinking that the current Pentium line is the same as the old Pentium 4 line. There is not enough faces and palms for this. Gah!

    A colleague mentioned there's been an update to the documentation (I'm reading the 2012 version). Hopefully we can obtain a copy of it soon, but we need to specifically request it and then have a paper copy sent to us. But hey, maybe they updated the instructions to use Linux Ubuntu 12 now!


  • 🚽 Regular

    @polygeekery

    That's...special. Was it teamed maybe?


  • Grade A Premium Asshole

    @cursorkeys said in The Official Status Thread:

    @polygeekery

    That's...special. Was it teamed maybe?

    No, but it was a virtual interface that had been used in the past as a way to route traffic from one VLAN to another. The role release was done through the web interface. I am going to assume that the web interface bungled the config change.

    Or there is something else I am missing and did not think of. I am not too worried about it right now. The noise in the logs were errors due to DHCP requests and it not having a pool defined in that subnet. Due to recent changes in the network, the firewall no longer needs to do DHCP at all. So I killed all DHCP which will also get rid of the noise in the logs.

    We can sort out the other part of it sometime when we are on-site.


  • Notification Spam Recipient

    @polygeekery said in The Official Status Thread:

    a virtual interface that had been used in the past as a way to route traffic from one VLAN to another

    Did you rename it em9_DO_NOT_TOUCH ? ;)


Log in to reply