That's what globals do.

On Tuesday, 27 June 2017 20:32:12 UTC+10, Nupur Bansal wrote:
>
> It seems my global variables are being accessed concurrently by goroutines 
> and it is causing race condition.
>
> On Tuesday, June 27, 2017 at 3:05:03 PM UTC+5:30, Dave Cheney wrote:
>>
>> You have at least one data race in your program.
>>
>> https://golang.org/doc/go1.6#runtime
>> https://blog.golang.org/race-detector
>> https://golang.org/doc/articles/race_detector.html
>>
>> Looking at the panic message it looks like your 
>> findSupplierOrBuyer method is updating a map that is part of a User's 
>> Dashboard. As the parent method is called GetDashboard it is possible that 
>> you are sharing the same Dashboard value between mutliple goroutines. 
>>
>> Have a read of the links above, and start running your tests under the 
>> race detector; go test -race.
>>
>> On Tuesday, 27 June 2017 19:25:57 UTC+10, Nupur Bansal wrote:
>>>
>>> Top part says:
>>>
>>> fatal error: concurrent map writes
>>>
>>> goroutine 23673 [running]:
>>> runtime.throw(0x95fc7f, 0x15)
>>>     /usr/lib/go/src/runtime/panic.go:566 +0x95 fp=0xc420482708 
>>> sp=0xc4204826e8
>>> runtime.mapassign1(0x8cb4c0, 0xc4200e8e10, 0xc420482a88, 0xc420482a68)
>>>     /usr/lib/go/src/runtime/hashmap.go:553 +0x2e1 fp=0xc4204827f0 
>>> sp=0xc420482708
>>> wservce/models.findSupplierOrBuyer(0xc42038812e, 0x6, 0x0, 0x0, 
>>> 0xc42073e270, 0xf, 0xc420388123, 0x2, 0xba9ac0, 0xc42051c800, ...)
>>>     /home/nupur/golang/code/src/wservce/models/userDashboard.go:153 
>>> +0xc07 fp=0xc420482e80 sp=0xc4204827f0
>>> wservce/models.(*UserDashboard).GetDashboardDetails(0xc4203622a0, 
>>> 0xba9ac0, 0xc42051c800, 0xc42073e270, 0xf, 0x1, 0x125)
>>>     /home/nupur/golang/code/src/wservce/models/userDashboard.go:75 
>>> +0x14d fp=0xc420483040 sp=0xc420482e80
>>> wservce/controllers.Request.Dashboard(0xba9ac0, 0xc42051c800, 
>>> 0xc42073e180, 0xf, 0xc4203880f4, 0x40, 0xc4203880f0, 0x3, 0xc42073e270, 
>>> 0xf, ...)
>>>     /home/nupur/golang/code/src/wservce/controllers/users.go:172 +0x106c 
>>> fp=0xc420483410 sp=0xc420483040
>>> wservce/controllers.(*Request).Dashboard(0xc4202b8660, 0x0, 0x0)
>>>     <autogenerated>:5 +0x6e fp=0xc420483488 sp=0xc420483410
>>> runtime.call32(0xc4204d09c0, 0xc4205aa060, 0xc420588960, 0x800000018)
>>>
>>>
>>> On Tuesday, June 27, 2017 at 10:04:06 AM UTC+5:30, Dave Cheney wrote:
>>>>
>>>> Your stacktrace is truncated; the crucial part appears at the top of 
>>>> the output starting with 
>>>>
>>>> panic:
>>>>
>>>> or 
>>>>
>>>> runtime error:
>>>>
>>>> On Tuesday, 27 June 2017 14:31:57 UTC+10, Nupur Bansal wrote:
>>>>>
>>>>> When a load test is run on my service, after about 1000-1500 requests, 
>>>>> I get this stack trace. I am unable to resolve why it was generated.
>>>>> Yes, the output from the service is passed to an output channel and is 
>>>>> read through that channel. 
>>>>>
>>>>> I am doing something like this:
>>>>> package main
>>>>>
>>>>> import (
>>>>>     "fmt"
>>>>>     //"time"
>>>>> )
>>>>>
>>>>> func myFunc(done chan string) {
>>>>>     // Doing something in parallel
>>>>>     for i := 0; i < 10; i++ {
>>>>>         fmt.Println(i)
>>>>>     }
>>>>>     fmt.Println("Hey! I do useless stuff!")
>>>>>     done <- "I'm done!" // We send a message on the channel
>>>>> }
>>>>>
>>>>> func main() {
>>>>>     done1 := make(chan string)
>>>>>     done2 := make(chan string)
>>>>>     go myFunc(done1)
>>>>>     go myFunc2(done2)
>>>>>     msg := <-done1
>>>>>     msg2 := <-done2
>>>>>     fmt.Println(msg, msg2)
>>>>>
>>>>>     fmt.Println("Message received, you were indeed useless..")
>>>>> }
>>>>>
>>>>> func myFunc2(done chan string) {
>>>>>     // Doing something in parallel
>>>>>     for i := 10; i < 20; i++ {
>>>>>         fmt.Println(i)
>>>>>     }
>>>>>     fmt.Println("Hey! Baby!")
>>>>>     done <- "I'm again done!" // We send a message on the channel
>>>>> }
>>>>>
>>>>>
>>>>> On Thursday, June 22, 2017 at 8:09:33 PM UTC+5:30, Ian Lance Taylor 
>>>>> wrote:
>>>>>>
>>>>>> On Wed, Jun 21, 2017 at 10:49 PM,  <nupur....@indiamart.com> wrote: 
>>>>>> > 
>>>>>> > I am getting the following error when my services are being load 
>>>>>> tested. 
>>>>>> > I am new to goroutines and may be missing something in my 
>>>>>> implementation. 
>>>>>> > My service on being hit is running 2 goroutines that call an http 
>>>>>> request. 
>>>>>> > After both return , the response from both http requests is 
>>>>>> returned into a 
>>>>>> > channel. 
>>>>>> > What could be the possible reason of the following stack trace of 
>>>>>> error?? 
>>>>>>
>>>>>> What caused the stack trace?  I would have expected to see something 
>>>>>> at the start saying why it was generated, but I didn't. 
>>>>>>
>>>>>> At first glance it looks like you have a deadlock somewhere.  Is 
>>>>>> there 
>>>>>> something reading from the channels? 
>>>>>>
>>>>>> Ian 
>>>>>>
>>>>>

-- 
You received this message because you are subscribed to the Google Groups 
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to golang-nuts+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to