Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[oldark]

I've been having crashes in various fortresses the past couple of days. I'm using the gold studded with patches download on windows7.  The tileset I'm using is Masterwork-Obsidian walls. 

The bottom of my error log is

Code: [Select]

*** Error(s) found in the file "data/save/region1/raw/objects/entity_default.txt"
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
MOUNTAIN:Unrecognized Entity Token: REACTION
path fail: goblin,7,26,173 -> 38,47,173: Id #6346:Path Goal Seek Station:Station Maraude Target at 39,47,173


[splinterz]

@splinterz, I changed the indicated section, now im getting a different error.

Code: [Select]

...\Dwarf Fortress\hack\scripts/fullheal.lua:73: Cannot write field vector (body_layer_satus).0: complex object.
stack traceback:
      [C]: in function '__newindex'
      ...f-fortress-gold\Dwarf Fortress\hack\scripts/fullheal.lua:73: in main chunk
     (...(tail calls...)


ah yeah, my mistake. as Putnam mentioned, it's a different structure. this seems to work:

Code: [Select]

v=unit.body.components
for i=0,#v.layer_status - 1,1 do
for j=0,#v.layer_status-1,1 do -- severed, leaking layers (Urist Da Vinci)
v.layer_status[j].gone = 0
v.layer_status[j].leaking = 0
end
v.layer_wound_area[i] = 0 -- wound contact areas (Urist Da Vinci)
v.layer_cut_fraction[i] = 0 -- 100*surface percentage of cuts/fractures on the body part layer (Urist Da Vinci)
v.layer_dent_fraction[i] = 0 -- 100*surface percentage of dents on the body part layer (Urist Da Vinci)
v.layer_effect_fraction[i] = 0 -- 100*surface percentage of "effects" on the body part layer (Urist Da Vinci)
end


[lcy03406]

"Sell oil for 500 shillings" consumes the container.
And also takes magma as oil.
And consumes the tank.

[blapnk]

So we can sell magma, and not just by dumping it on elf traders?

[lcy03406]

So we can sell magma, and not just by dumping it on elf traders?

Yes but for dwarves an iron tank is more valuable than silver coins.

[There Is No Vic]

Yes, I thought the same, mostly because I have never seen them actually dig. Now, that in itself is not that big of a deal, because I dont play much, but I never read any "whats happening in your fort" posts about things that dug into a fort either. In short, not a single report since I added this.

Best to remove it, and bug expwnent some more. I know it should work, he is professional enough not to release anything untested, but obviously there are some flaws.

I have seen 2 orc raids designate a downward stair for digging outside my fort. I killed the ambushers after a couple of months, so nothing came of it.

[Jakob]

-Woops wrong place-

[urmane]

MW 4j, on linux - have a long running fort again, got tired of not having workflow, so I decided to start debugging.  My startup dfhack output I've now fixed down to this:

Spoiler (click to show/hide)

$ ./dfhack
Loading bindings from data/init/interface.txt
New window size: 1280x400
Font size: 16x16
Resizing grid to 80x25
Resizing font to 16x16

Resetting textures
./hack/lua/dfhack.lua:223: attempt to index field 'random' (a nil value)
stack traceback:
        ./hack/lua/dfhack.lua:223: in main chunk
        [C]: in ?
automaterial is not a recognized command.
autoSyndrome is enabled
Monitoring enabled: all
Could not save the status.
Valid plant IDs:
manipulator is not a recognized command.
Overlay enabled
search is not a recognized command.
Map is not available!
syndromeTrigger is not a recognized command.
attempt to call a nil value
stack traceback:
Monitoring enabled: all
attempt to call a nil value
stack traceback:
Enabled tweak stable-cursor
Enabled tweak patrol-duty
Enabled tweak readable-build-plate
Enabled tweak stable-temp
Enabled tweak stable-temp
Enabled tweak fast-heat
Enabled tweak fast-heat
Enabled tweak fast-heat
Enabled tweak fix-dimensions
Enabled tweak fix-dimensions
Enabled tweak fix-dimensions
Enabled tweak fix-dimensions
Enabled tweak fix-dimensions
Enabled tweak advmode-contained
Enabled tweak fast-trade
Enabled tweak fast-trade
Enabled tweak military-stable-assign
Enabled tweak military-color-assigned
Enabled tweak military-training
Enabled tweak military-training
Enabled tweak military-training
Enabled tweak military-training
attempt to call a nil value
stack traceback:
attempt to call a nil value
stack traceback:
attempt to call a nil value
stack traceback:
attempt to call a nil value
stack traceback:
FixGrowth: Running
DFHack is ready. Have a nice day!
Type in '?' or 'help' for general help, 'ls' to see all commands.
[DFHack]#

To get this far, I've removed all lines containing "diggingInvaders", commented out all the binpatches and the Gnome and Warlock scripts (they all just threw the lua error "attempt to call a nil value" anyway), and changed this:

Spoiler (click to show/hide)

enable automaterial
enable autoSyndrome
enable autotrade
enable buildingplan
enable dwarfmonitor
enable fix-armory
enable getplants
enable manipulator
enable mousequery
enable resume
enable search
enable zone

to this:

Spoiler (click to show/hide)

automaterial
autoSyndrome
autotrade
buildingplan
dwarfmonitor enable all
fix-armory enable
getplants
manipulator
mousequery plugin enabled
resume show
search
zone

(which doesn't make any sense to me - "enable" is reported as not valid, I don't understand how these are working for anyone if that's true ...)

The "not a recognized command" lines confuse me - each one is a plugin, for which there is a .so in hack/plugins (except for syndromeTrigger, which explains that one).  No other errors associated with those ...

The remaining "attempt to call a nil value" errors come from the following lines:
(first two - not sure)
itemsyndrome contaminantsOn enable
itemsyndrome enable
soundsense-season (confusing, as soundSense notices the seasons just fine)
fix/cloth-stockpile enable

Any thoughts the audience has would be greatly appreciated.  Now that I've typed all this, I'm wondering if my UtilityPlugins_v0.35-Windows-0.34.11.r3.zip is too old ...

[urmane]

Utility Plugins updated to v0.43, more working now.  MDF ships with the Windows dlls - for linux one has to obtain the .so shared libraries separately, and mine were dated.

The workflow plugin, it turns out, throws an error in stderr.log about GLIBCXX_3.4.15.  The linux version of DF comes with a libs/libstdc++.so.6 that's too old - moving/renaming that allows the stocks.so plugin to load.  Yay, I've got workflow again!

Still getting a lot of script/plugin failures ("attempt to call a nil value") - I must be doing something wrong ... is MDF using something other than basic dfhack-0.34.11-r3 ... ?

Also this:

Code: [Select]

./hack/lua/dfhack.lua:223: attempt to index field 'random' (a nil value)
stack traceback:
        ./hack/lua/dfhack.lua:223: in main chunk
        [C]: in ?
Line 223 is this:

Code: [Select]

dfhack.random.__index = dfhack.random
Despite all this, the game itself continues to run just fine.

[Meph]

I use the inofficial r4 dfhack release, necessary for several new scripts.

[urmane]

I use the inofficial r4 dfhack release, necessary for several new scripts.

Well, that explain a lot :-)

I'll see if I can compile it for linux, and while I'm at it I'll bite the bullet and layman qt5 to get an up-to-date DT ...

[Repseki]

Just had two diseased migrants show up at my fort, even though I turned diseases off. Not entirely sure they are actually diseased, as things are paused at the moment, but they definitely seem to be.

The first is covered in blisters, with Therapist showing a smallpox buff. With the extra fun of almost definitely being a Vampire.

The second shows smallpox and bubonic plague, while also having pock-marked plaguebearer tagged onto their name.

Just kinda weird.

[Tenderroast]

Repseki, were the two diseased migrants part of the 1st to migration waves?

[lcy03406]

My rockmage is not big enough to dual wield great weapons. I don't know if it was intended or a bug. Her profile says "She is weak. She is weak and short." and Dwarf Therapist reports her size is only 52950. Is it too small to dual wield?
And also her profession is "rockmage hunter *earthkin*". I think it should be "rockmage hunter *rockmage*".
And I failed 3 times to train her in "marster the school of earth". "Failed" means her took all reagents and begun to work, and the reagents disappeared, an announcement showed up but her didn't transform.

[splinterz]

My rockmage is not big enough to dual wield great weapons. I don't know if it was intended or a bug. Her profile says "She is weak. She is weak and short." and Dwarf Therapist reports her size is only 52950. Is it too small to dual wield?
And also her profession is "rockmage hunter *earthkin*". I think it should be "rockmage hunter *rockmage*".
And I failed 3 times to train her in "marster the school of earth". "Failed" means her took all reagents and begun to work, and the reagents disappeared, an announcement showed up but her didn't transform.

the unit's size and their relation to weapons is pretty buggy in DF in general. from all the testing i've done, it uses the default adult size for the weapon checks, regardless of the dwarf's actual size. the only way to know for sure is to try and assign them the weapons and see if they'll pick them up.

there's some issue with the transformations that have been reported. some other people have reported that the rock mages' shield blocks the transformation i think? it could be that.