Adch++ goes down

Bug #1203313 reported by Janne
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
ADCH++
New
Undecided
Unassigned

Bug Description

Hub goes down sometimes for some reason.why we do not know, it may be that we have too many users in the Hub we have 600-1100 users.we can not see anything in the log
any idea why it goes down??

ADCH++ 2.10.1 (r646) Release

Janne

Revision history for this message
poy (poy) wrote :

is this a new problem with version 2.10.1 or has it happened before?

check the available CPU / memory on the server to see if the hub is hitting any limits.

affects: adchpp-ptokax → adchpp
Revision history for this message
Janne (barajag) wrote :

this problem, we have had a long time even in the old.
sometimes the hub roll 3-5 days without going down. then it can go down 2-3 times in a day

no limits in the computer
4*3.8Ghz
8GB memory

only adch++ runs on that computer and Teamviewer

Janne

Revision history for this message
Janne (barajag) wrote :

hub has been down two times today
maby some of the script causing the problem?

i dont know

Janne

Revision history for this message
poy (poy) wrote :

ADCH++ hubs are running fine for many people so this is likely to be caused by a script, indeed. i suggest half-splitting: enable half of the scripts, disable the other half and see if it crashes; then keep going until you can determine which script(s) is the culprit.

Revision history for this message
Janne (barajag) wrote :

you dont think it has to do whit the number users ?

Janne

Revision history for this message
Janne (barajag) wrote :

I've discovered another thing now if you ban a user. he will not be kicked from the hub
it should work that way?

Janne

Revision history for this message
Janne (barajag) wrote :

would you check this
Adch++ 2.11.0 r649
bridge 112

Revision history for this message
Janne (barajag) wrote :

one more file

Revision history for this message
poy (poy) wrote :

are you using a chat room script? this seems to be the same error as in bug 1155321.

not much to be found in the crash logs unfortunately, except that they are likely caused by Lua scripts. i suggest disabling some to see if the situation improves.

Revision history for this message
Janne (barajag) wrote :

no not any chat room script. we vill test disabling some script the next time it goes down

Revision history for this message
Janne (barajag) wrote :
Download full text (5.0 KiB)

we have a script that might cause this problem

RatioChecker.lua

--[[

 HSS Check 1.0b LUA 5.1x [Strict] [API 2]

 Hub Share Slot Check
 By Mutor 12/25/07

 Provides limits per profile for:
  -Minimum share
  -Maximum hubs
  -Minimum slot:hub ratio
 Option for disconnect,kick or ban
 Messages user before removal
 Reports to: Nick, Ops or All
 Checks for myinfo changes after login
 Protect by nick option
 Action taken if MyINFO is invalid

 +Changes from 1.0 12/27/07
  +Added option to use hub/slot ratio or max slot setting
  +Added clean hub on script start
  ~Changed MyINFO checking [checked only after connection]

]]

-- Reporting function
-- 0 = Disabled 1 = SendToNick 2 = SendToOps, 3 = SendToAll
local Report = 3
-- If Report = 0, then set Nick
local OpNick = ""--"Mutor"
-- Name for Bot
local Bot = SetMan.GetString(21)
-- Penalty for offenders 1=Disconnect, 2 = Kick[default tempban], 3 = Ban
local Penalty = 1
-- Set Your Profiles
-- [#] = {MinShare in GB,MaxHubs,*Slot/Hub Ratio OR Max Slots},
local Profiles = {
 [-1] = {30960,0,0},
 [0] = {0,0,0},
 [1] = {0,0,0},
 [2] = {0,0,0},
 [3] = {0,0,0},
 [4] = {0,0,0},
 [5] = {512,0,0},
 [6] = {1024,0,0},
 [7] = {10,0,0},
 }
-- *Use slot count as a ratio? [3rd value in the Profiles array] true/false
-- *If set to false slot count is max slot limit set Profiles array accordingly
local Ratio = false
local SafeNicks = {
 ["[10Mbit]Ben"] = true,
 ["WebReg"] = true,
 ["ruler"] = true,
 ["123Rizzo"] = true,
 }

OnStartup = function()
 for _,user in ipairs(Core.GetOnlineUsers()) do
  CheckUser(user)
 end
 collectgarbage("collect")
end

UserConnected = function(user)
 CheckUser(user)
end
OpConnected,RegConnected = UserConnected,UserConnected

MyINFOArrival = function(user,data)
 if Core.GetUserValue(user,9) then
  CheckUser(user)
 end
end

CheckUser = function(user)
 if not SafeNicks[user.sNick] then
  local t = {}
  for i = 16, 21 do
   if i < 18 or i > 20 then
    local x = Core.GetUserValue(user,i)
    if x then
     table.insert(t,tonumber(x))
    end
   end
  end
  if #t == 3 then
   local tab = Profiles[user.iProfile]
   if tab then
    local rsn
    local p,m,z,x = GetProf(user.iProfile),"",1024^3
    if tab[1] > 0 and t[1] < (tab[1]* z) then
     rsn = "Under minimum share limit"
     m,x = FmtSz((tab[1]* z) - t[1]).." short of minimum share of "..
     FmtSz(tab[1] * z).." for "..p.."s.",true
    elseif tab[2] > 0 and t[2] > tab[2] then
     rsn = "Over maximum hub limit."
     m,x = t[2] - tab[2].." too many hubs (Max: "..tostring(tab[2])..
     " ) for "..p.."s.",true
    else
     if Ratio then
      --if tab[3]> 0 and (t[3]/t[2] < tab[3]*t[2]) then
      if t[2] < 1 and t[3] < 1 or tab[3]> 0 and (t[3]/t[2] < tab[3]*t[2]) then
       rsn = "Under minimum slot:hub ratio"
       local hs,cs = tab[3]*t[2],math.ceil((tab[3]*t[2]) - t[3])
       m,x = tostring(cs).." too few slots for "..p.."s. "..
       "You're logged on to "..tostring(t[2]).." hubs which requires "..
       "you to have "..tostring(hs).." slots.",true
      end
     else
      if tab[3] > 0 and t[3] > tab[3] then
       rsn = "Over maximum slot allowance"
       local hs,cs = tab[3],t[3] - tab[3]
       m,x = tostring(cs)...

Read more...

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.