Re: Delay process bug

[Jim Crate via 4D_Tech]

I would like to emphasize this. It’s likely processes delayed with DELAY 
PROCESS are woken up by some internal process that ends up getting stopped 
somehow. The bug is most likely not in DELAY PROCESS, but the internal process 
that (among many other things) checks the list of delayed processes to see when 
they should wake up.

Interestingly, people have noted that a form running SET TIMER does not seem to 
be affected. We don’t know what else is affected by that core internal process 
getting stopped other than the user interface. Backups still run, at least.

Jim Crate


> On Jun 18, 2018, at 5:53 PM, Jeffrey Kain via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> It seems like a big clue is that the 4D Server user interface stops updating 
> when this happens. Like the internal clock/heartbeat/metronome of 4D Server 
> just stops, and DELAY PROCESS comes along for the ride.
> 
> --
> Jeffrey Kain
> jeffrey.k...@gmail.com
> 
>> On Jun 18, 2018, at 8:36 PM, Justin Carr via 4D_Tech <4d_tech@lists.4d.com> 
>> wrote:
>> 
>> Same as Cannon. One background process that does quite a lot. Others that 
>> just wake up, check something and go back to sleep. There doesn't seem to be 
>> any rhyme or reason to it which is why it's so difficult to reproduce. I'm 
>> not sure how it *couldn't* be a problem with DELAY PROCESS.
> 
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Magnus Torell via 4D_Tech]

Just for the record I can confirm that I have a mac server
with a stored procedure that works like beneath, where I have seen
this freezing state.

Repeat
.
Find potential things to do
Do it if any
.
Delay Process(Current Process;60*60)  // Delay 60 sec
Until (stopCallFromOutside)

This is a v15R5 server and I think I have had this method to freeze
2 or 3 times times over a period of 1 and a half year.
There is no Pause or resume process involved and it is pretty clean.
On these occasions I could not stop communicate with this process
and it did stop/freeze. Beside that server was doing fine. Did not check
for interface update. Only way to resolve it was to restart.

Server is running compiled and I "upgrade" frequently i.e. restarts every
week.
If some say it was introduced in v16 it may to have entered in the
R5 release of v15R5

Best
Magnus Torell



--
Sent from: http://4d.1045681.n5.nabble.com/4D-Tech-f1376241.html
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bernd Fröhlich via 4D_Tech]

Jeffrey Kain:

> It seems like a big clue is that the 4D Server user interface stops updating 
> when this happens. Like the internal clock/heartbeat/metronome of 4D Server 
> just stops, and DELAY PROCESS comes along for the ride.

Server might be a clue indeed. It happened only on the Server over here.
Although the server interface was still working when I encountered the bug.

How about the others that have been bitten?
Anyone here where the bug was encountered on a client (or standalone)?

Greetings from Germany,
Bernd Fröhlich
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jeffrey Kain via 4D_Tech]

It seems like a big clue is that the 4D Server user interface stops updating 
when this happens. Like the internal clock/heartbeat/metronome of 4D Server 
just stops, and DELAY PROCESS comes along for the ride.

--
Jeffrey Kain
jeffrey.k...@gmail.com

> On Jun 18, 2018, at 8:36 PM, Justin Carr via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> Same as Cannon. One background process that does quite a lot. Others that 
> just wake up, check something and go back to sleep. There doesn't seem to be 
> any rhyme or reason to it which is why it's so difficult to reproduce. I'm 
> not sure how it *couldn't* be a problem with DELAY PROCESS.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jeffrey Kain via 4D_Tech]

This is a new bug in version 16.  Lots of experience it but nobody can 
reproduce it.

> On Jun 18, 2018, at 5:02 PM, Chuck Miller via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> i use delay process for years with no issues, however, I do not think I ever 
> delay for more than 15 minutes.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Chuck Miller via 4D_Tech]

i use delay process for years with no issues, however, I do not think I ever 
delay for more than 15 minutes.

Chuck

 Chuck Miller Voice: (617) 739-0306
 Informed Solutions, Inc. Fax: (617) 232-1064   
 mailto:cjmillerinformed-solutions.com 
 Brookline, MA 02446 USA Registered 4D Developer
   Providers of 4D and Sybase connectivity
  http://www.informed-solutions.com  

This message and any attached documents contain information which may be 
confidential, subject to privilege or exempt from disclosure under applicable 
law.  These materials are intended only for the use of the intended recipient. 
If you are not the intended recipient of this transmission, you are hereby 
notified that any distribution, disclosure, printing, copying, storage, 
modification or the taking of any action in reliance upon this transmission is 
strictly prohibited.  Delivery of this message to any person other than the 
intended recipient shall not compromise or waive such confidentiality, 
privilege or exemption from disclosure as to this communication. 

> On Jun 18, 2018, at 2:39 AM, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> Justin,
> I do use DELAY PROCESS but as I said it's only for a minute or two.
> And like I said maybe I've simply been lucky. As my business mentor used to
> say, "sometimes it's better to be lucky than smart."

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Cannon Smith via 4D_Tech]

I can’t answer this for Justin, but we have both kinds hitting this bug. Some 
are rather complicated methods that potentially do lots of things. But several 
are very simple—just a few lines of code to check whatever it needs to check 
and then go back to sleep.

--
Cannon.Smith
Synergy Farm Solutions Inc.
Hill Spring, AB Canada
403-626-3236




> On Jun 18, 2018, at 9:32 AM, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> I bet the bug is actually someplace else and is manifesting in this DELAY
> PROCESS bit. Which doesn't help you find it at the moment, though. Is the
> method running this process one of those big, complicated, gnarly types
> with all sorts of convoluted branchings and calls to numerous plugins and
> dependencies on lots of IP vars?

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jeffrey Kain via 4D_Tech]

Our development server stopped updating its user interface today. Same symptoms 
as the DELAY PROCESS bug - the admin graphs like CPU usage stop, the list of 
connected users goes out of date, uptime calculations reflect the time the 
internal clock stopped.

Normally we start getting alerts on our production servers because we're in a 
mirrored enviornment and the journal files start stacking up.  Today was the 
first time that our development server, running interpreted with no mirror 
code, just stopped updating.

> On Jun 18, 2018, at 11:32 AM, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> Justin,
> Yeah, after I responded to you I read Miyako's comment - and your reply.
> It's an incredibly frustrating place to be when you do everything right and
> the wrong thing still happens. Unlucky.
> 
> I bet the bug is actually someplace else and is manifesting in this DELAY
> PROCESS bit. Which doesn't help you find it at the moment, though. Is the
> method running this process one of those big, complicated, gnarly types
> with all sorts of convoluted branchings and calls to numerous plugins and
> dependencies on lots of IP vars?

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Kirk Brooks via 4D_Tech]

Justin,
Yeah, after I responded to you I read Miyako's comment - and your reply.
It's an incredibly frustrating place to be when you do everything right and
the wrong thing still happens. Unlucky.

I bet the bug is actually someplace else and is manifesting in this DELAY
PROCESS bit. Which doesn't help you find it at the moment, though. Is the
method running this process one of those big, complicated, gnarly types
with all sorts of convoluted branchings and calls to numerous plugins and
dependencies on lots of IP vars?

On Mon, Jun 18, 2018 at 12:18 AM Justin Carr via 4D_Tech <
4d_tech@lists.4d.com> wrote:

> On 18 Jun 2018, at 4:39 pm, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com>
> wrote:
> >
> > Justin,
> > I do use DELAY PROCESS but as I said it's only for a minute or two.
> > And like I said maybe I've simply been lucky. As my business mentor used
> to
> > say, "sometimes it's better to be lucky than smart."
> >
>
> Hi Kirk
>
> We are only delaying for 10 seconds but our users were still experiencing
> this problem from time to time. We have worked around it now using a method
> similar to those already posted on this thread but it would sure be nice to
> once again be able to trust a delayed process to wake up when it's supposed
> to!
>


-- 
Kirk Brooks
San Francisco, CA
===

*We go vote - they go home*
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Justin Carr via 4D_Tech]

On 18 Jun 2018, at 4:39 pm, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com> 
wrote:
> 
> Justin,
> I do use DELAY PROCESS but as I said it's only for a minute or two.
> And like I said maybe I've simply been lucky. As my business mentor used to
> say, "sometimes it's better to be lucky than smart."
> 

Hi Kirk

We are only delaying for 10 seconds but our users were still experiencing this 
problem from time to time. We have worked around it now using a method similar 
to those already posted on this thread but it would sure be nice to once again 
be able to trust a delayed process to wake up when it's supposed to!

J
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Kirk Brooks via 4D_Tech]

Justin,
I do use DELAY PROCESS but as I said it's only for a minute or two.
And like I said maybe I've simply been lucky. As my business mentor used to
say, "sometimes it's better to be lucky than smart."

On Sun, Jun 17, 2018 at 7:45 PM Justin Carr via 4D_Tech <
4d_tech@lists.4d.com> wrote:

> On 16 Jun 2018, at 12:20 am, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com>
> wrote:
> >
> > I have never had an issue with DELAY PROCESS. Maybe I'm lucky. I don't
> > think I've ever used it to delay something more than a matter of seconds,
> > though. It would never have occurred to me to DELAY PROCESS for minutes
> or
> > hours.
>
> 
>
> > The other tool I have for such situations is a really simple stored
> method
> > launcher. It requires 1 table, a background loop (which delays 1 minute)
> > and the idea is to create a record for each method that I want to run at
> > some interval.
>
> Hi Kirk
>
> Just curious... how do you delay your background loop for a minute without
> using DELAY PROCESS? If you're using it for this then you'll be potentially
> affected by this bug and, as you pointed out, just lucky it hasn't bitten
> you yet.
>

-- 
Kirk Brooks
San Francisco, CA
===

*We go vote - they go home*
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bernd Fröhlich via 4D_Tech]

Keisuke Miyako:

> it's not that the DELAY P. has a problem resuming,
> the problem manifests itself when the codes pauses a process that is already 
> in "delay" status.
> a double resume is required to cancel a delay-double combo.

That would be another bug then.

Over here it was a simple loop with Delay Process that stopped working after 
some time. It just did not wake up anymore. No Pause Process involved.

Greetings from Germany,
Bernd Fröhlich
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Justin Carr via 4D_Tech]

On 18 Jun 2018, at 1:31 pm, Keisuke Miyako via 4D_Tech <4d_tech@lists.4d.com> 
wrote:
> 
> according to the bug description for ACI0098368 ("Delayed processes do not 
> wake up if paused/resumed while already delayed"),
> 
> it's not that the DELAY P. has a problem resuming,
> the problem manifests itself when the codes pauses a process that is already 
> in "delay" status.
> a double resume is required to cancel a delay-double combo.
> 
>> In v16 the runtime explorer shows the following:
>> 
>> - start process: Executing
>> - delay process (3600 ticks): Delayed
>> - pause process (within the delay time): Delayed
>> - resume process (within the delay time): Paused (!)
>> - resume process or delay process (0 ticks): Executing
>> 
>> So in v16, there is an additional RESUME PROCESS that is required to kick 
>> the process back into gear.
> 
> so as long as you use pause & resume, or delay (with no interference), 
> everything should be OK.
> 
> but mixing delay, pause, resume in a complex sequence could mean trouble.
> the good news is that the bug described above is on course to be fixed.

I think that is a different bug from the one many of us are experiencing. There 
is no PAUSE PROCESS called on the delayed process. It simply doesn't wake up at 
the allocated time.
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Keisuke Miyako via 4D_Tech]

according to the bug description for ACI0098368 ("Delayed processes do not wake 
up if paused/resumed while already delayed"),

it's not that the DELAY P. has a problem resuming,
the problem manifests itself when the codes pauses a process that is already in 
"delay" status.
a double resume is required to cancel a delay-double combo.

> In v16 the runtime explorer shows the following:
>
> - start process: Executing
> - delay process (3600 ticks): Delayed
> - pause process (within the delay time): Delayed
> - resume process (within the delay time): Paused (!)
> - resume process or delay process (0 ticks): Executing
>
> So in v16, there is an additional RESUME PROCESS that is required to kick the 
> process back into gear.

so as long as you use pause & resume, or delay (with no interference), 
everything should be OK.

but mixing delay, pause, resume in a complex sequence could mean trouble.
the good news is that the bug described above is on course to be fixed.



**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Justin Carr via 4D_Tech]

On 16 Jun 2018, at 12:20 am, Kirk Brooks via 4D_Tech <4d_tech@lists.4d.com> 
wrote:
> 
> I have never had an issue with DELAY PROCESS. Maybe I'm lucky. I don't
> think I've ever used it to delay something more than a matter of seconds,
> though. It would never have occurred to me to DELAY PROCESS for minutes or
> hours.



> The other tool I have for such situations is a really simple stored method
> launcher. It requires 1 table, a background loop (which delays 1 minute)
> and the idea is to create a record for each method that I want to run at
> some interval.

Hi Kirk

Just curious... how do you delay your background loop for a minute without 
using DELAY PROCESS? If you're using it for this then you'll be potentially 
affected by this bug and, as you pointed out, just lucky it hasn't bitten you 
yet.

J
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Kirk Brooks via 4D_Tech]

I have never had an issue with DELAY PROCESS. Maybe I'm lucky. I don't
think I've ever used it to delay something more than a matter of seconds,
though. It would never have occurred to me to DELAY PROCESS for minutes or
hours.

If I have some process I only want one instance of, maybe processing
records created by imports from various, simultaneous sources, I use this
outline:

// UniqueProcessMethod (some parameter)
$processName:=Current method name
Case of

:(Count parameters=1)

$pid:=Process number($processName)

If($pid=0)  // process doesn't exist

$pid:=EXECUTE ON SERVER(Current method name; 1024*1024; $processName;
$1;Current process)

End if

:(Count parameters=2)

//  do whatever here

End case


This works consistently if called from a client process or server process.
I prefer making $1 an object but the approach works just as well if you
have more than one parameter. The key is adding the current process
(sometimes useful) to distinguish between the call to start the process and
just checking data.

The other tool I have for such situations is a really simple stored method
launcher. It requires 1 table, a background loop (which delays 1 minute)
and the idea is to create a record for each method that I want to run at
some interval. I'm sure a lot of you already have such a thing in place.
This is what I use in place of long delays. Like for daily reports or
maintenance methods.

On Fri, Jun 15, 2018 at 12:04 AM Bernd Fröhlich via 4D_Tech <
4d_tech@lists.4d.com> wrote:

> Bryan Green:
>
> > Actually, that's exactly what I meant -- short-term (5 minutes, 10
> minutes, etc.) kind of delay using DELAY PROCESS is more stable than
> something like 12 hours, 1 day, etc. using DELAY PROCESS.
>
> If by "more stable" you mean "hangs only every few weeks" that may be
> right.
> I still would not use it at the moment.
> Until it stopped beeing reliable I used it to delay a process by 10
> seconds.
>
> Greetings from Germany,
> Bernd Fröhlich
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **



-- 
Kirk Brooks
San Francisco, CA
===

*We go vote - they go home*
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bernd Fröhlich via 4D_Tech]

Bryan Green:

> Actually, that's exactly what I meant -- short-term (5 minutes, 10 minutes, 
> etc.) kind of delay using DELAY PROCESS is more stable than something like 12 
> hours, 1 day, etc. using DELAY PROCESS.

If by "more stable" you mean "hangs only every few weeks" that may be right.
I still would not use it at the moment.
Until it stopped beeing reliable I used it to delay a process by 10 seconds.

Greetings from Germany,
Bernd Fröhlich
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bryan Green via 4D_Tech]

> On Jun 13, 2018, at 14:28, Chuck Miller via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> What does that mean? If you have a cron job process running on server that 
> wakes every 5 minutes to check and launch other processes. Is that what you 
> consider short term scheduling?

Actually, that's exactly what I meant -- short-term (5 minutes, 10 minutes, 
etc.) kind of delay using DELAY PROCESS is more stable than something like 12 
hours, 1 day, etc. using DELAY PROCESS.

You're in my head, Chuck!

-- bag


**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Chuck Miller via 4D_Tech]

What does that mean? If you have a cron job process running on server that 
wakes every 5 minutes to check and launch other processes. Is that what you 
consider short term scheduling?

Regards

Chuck

 Chuck Miller Voice: (617) 739-0306
 Informed Solutions, Inc. Fax: (617) 232-1064   
 mailto:cjmillerinformed-solutions.com 
 Brookline, MA 02446 USA Registered 4D Developer
   Providers of 4D and Sybase connectivity
  http://www.informed-solutions.com  

This message and any attached documents contain information which may be 
confidential, subject to privilege or exempt from disclosure under applicable 
law.  These materials are intended only for the use of the intended recipient. 
If you are not the intended recipient of this transmission, you are hereby 
notified that any distribution, disclosure, printing, copying, storage, 
modification or the taking of any action in reliance upon this transmission is 
strictly prohibited.  Delivery of this message to any person other than the 
intended recipient shall not compromise or waive such confidentiality, 
privilege or exemption from disclosure as to this communication. 

> On Jun 12, 2018, at 5:07 PM, Bryan Green via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> That happened to me once in an earlier version of 4D. Long story short, DELAY 
> PROCESS is meant for short-term scheduling, not long-term process management. 
> Even so, it’s good to know in advance. 
> 
> -- bag

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bryan Green via 4D_Tech]

That happened to me once in an earlier version of 4D. Long story short, DELAY 
PROCESS is meant for short-term scheduling, not long-term process management. 
Even so, it’s good to know in advance. 

-- bag

Sent from my mobile

On Jun 12, 2018, at 13:57, Dennis, Neil via 4D_Tech <4d_tech@lists.4d.com> 
wrote:

>> 32-bit long integer flips to negative, after 24 days.
> 
> You might be on to something 
> 
> Could pass this on to the 4D tech guys... there are several bugs already 
> reported for this.
> 
> Neil
> 
> 
> 
> Privacy Disclaimer: This message contains confidential information and is 
> intended only for the named addressee. If you are not the named addressee you 
> should not disseminate, distribute or copy this email. Please delete this 
> email from your system and notify the sender immediately by replying to this 
> email.  If you are not the intended recipient you are notified that 
> disclosing, copying, distributing or taking any action in reliance on the 
> contents of this information is strictly prohibited.
> 
> The Alternative Investments division of UMB Fund Services provides a full 
> range of services to hedge funds, funds of funds and private equity funds.  
> Any tax advice in this communication is not intended to be used, and cannot 
> be used, by a client or any other person or entity for the purpose of (a) 
> avoiding penalties that may be imposed on any taxpayer or (b) promoting, 
> marketing, or recommending to another party any matter addressed herein.
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Koen Van Hooreweghe via 4D_Tech]

Hi Jim,

Is there somehow PAUSE PROCESS/RESUME PROCESS involved?

I don't know if this has something to do with our issue, but I'd like to share 
a change in behavior (bug?) between v15 (and older) and v16 (and later).
Also struggling with a process not waking up after the delay time, I found 
pause process/resume process on a delayed process does not work as I would 
expect in 4D v16 and later.
Following scenario:

A process is running and is delayed for eg 1 minute. If the process is paused 
and resumed within that minute, the process won't wake up. Neiter immediately, 
neither after the delay time. Tested on Mac (v16 and 17 32/64). Another call to 
RESUME PROCESS will wake it immediately, another DELAY PROCESS will wake it 
after the new delay time.

This behavior is different in v15. This is what I do and what I see in the 
runtime explorer:
- start process: Executing
- delay process (3600 ticks): Delayed
- pause process (within the delay time): Delayed
- resume process (within the delay time): Executing

V16
- start process: Executing
- delay process (3600 ticks): Delayed
- pause process (within the delay time): Delayed
- resume process (within the delay time): Paused (!)

- resume process or delay process (0 ticks): Executing

I don't know what the standard behavior should be, but above is what I 
discovered. The work around is rather easy: when you want to resume a process, 
check if its status is paused or delayed. If paused: resume, if delayed: 
delay(0). Or don't use resume, but always use delay(0). Because a delayed or 
paused process will always wake up on delay(0).

HTH
Koen

> Op 12 jun. 2018, om 21:43 heeft Jim Crate via 4D_Tech <4d_tech@lists.4d.com> 
> het volgende geschreven:
> 
> Our server (macOS 10.11, 4D 16.3 HF3 64-bit) hit this bug again last week 
> after being up for about 22 days. There are 2 separate processes that run 
> periodically, both at the same time stopped waking from DELAY PROCESS. The 
> Server window also was not showing the 4D username in the user list, although 
> IP, machine name, and login date/time were displayed.




Compass bvba
Koen Van Hooreweghe
Kloosterstraat 65
9910 Knesselare
Belgium
tel +32 495 511.653

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Bernd Fröhlich via 4D_Tech]

Cannon Smith:

> We just released a new version of our software a week and a half ago. In this 
> version I added a palette window on the server which runs On Timer. I also 
> added code to most every background process which updates some information 
> every time it wakes up from a Delay Process call. The palette window checks 
> this information and, if it sees a process that has been asleep longer than 
> one minute, resumes the process and sends us a Slack message identifying the 
> server and process that was asleep. If the process does not get resumed, 
> another Slack message is sent so we can handle the issue manually before our 
> customer knows.
> 
> Interestingly, this mechanism was triggered on Sunday morning at 1:01 AM for 
> 9 servers. All 9 reported one or more stalled processes within a 60 second 
> time period.


I also have a palette window that runs On Timer.
So the workaround for the Delay Process bug was quite simple over here.
I replaced Delay Process with Pause Process (which is what Delay Process does 
anyway when the bug bites).
In On Timer I call Resume Process.
Never had any problems since I did it that way.

(Still this is only a workaround and the bug should of course be fixed by 4D. 
Until it is fixed Delay Process is a command that should be avoided.)

Greetings from Germany,
Bernd Fröhlich
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: Delay process bug

[Dennis, Neil via 4D_Tech]

I also did some digging our servers are patched on a monthly bases and 
restarted. So usually restart is few days over the 24.74 days. We have only had 
one instance of the delay process bug. So most of the time, even if it is a 32 
bit signed integer problem, we don't have an issue with the delay process 
except the one time.

Neil





Privacy Disclaimer: This message contains confidential information and is 
intended only for the named addressee. If you are not the named addressee you 
should not disseminate, distribute or copy this email. Please delete this email 
from your system and notify the sender immediately by replying to this email.  
If you are not the intended recipient you are notified that disclosing, 
copying, distributing or taking any action in reliance on the contents of this 
information is strictly prohibited.

The Alternative Investments division of UMB Fund Services provides a full range 
of services to hedge funds, funds of funds and private equity funds.  Any tax 
advice in this communication is not intended to be used, and cannot be used, by 
a client or any other person or entity for the purpose of (a) avoiding 
penalties that may be imposed on any taxpayer or (b) promoting, marketing, or 
recommending to another party any matter addressed herein.
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jim Crate via 4D_Tech]

I doublechecked the time the server was started until the last time the 
processes were delayed and it was actually 24.74 (2137700 seconds). The 
processes were delayed for 5 minutes. However, that time period + the delay is 
still 9,483,647 ms, or 2.63 hours, short of the 2147483647 max signed 32-bit 
integer.

Also many people have this occur after several months, so the DELAY PROCESS 
code must be able to accommodate the milliseconds rollover, at least in 
general. If it’s related to that, it must be an interesting corner case.

Jim Crate


> On Jun 12, 2018, at 1:33 PM, Keisuke Miyako via 4D_Tech 
> <4d_tech@lists.4d.com> wrote:
> 
> 22 days does sound suspicious; as if (no proof),
> the scheduler is testing if milliseconds "minus" process up-time is "greater 
> than" delay time...
> such logic would break when milliseconds expressed in signed 32-bit long 
> integer flips to negative, after 24 days.
> (milliseconds is counted since system boot, not 4D launch)
> 
>> 2018/06/13 4:43、Jim Crate via 4D_Tech <4d_tech@lists.4d.com> のメール：
>> Our server (macOS 10.11, 4D 16.3 HF3 64-bit) hit this bug again last week 
>> after being up for about 22 days.
> 
> 
> 
> 
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: Delay process bug

[Dennis, Neil via 4D_Tech]

> 32-bit long integer flips to negative, after 24 days.

You might be on to something 

Could pass this on to the 4D tech guys... there are several bugs already 
reported for this.

Neil



Privacy Disclaimer: This message contains confidential information and is 
intended only for the named addressee. If you are not the named addressee you 
should not disseminate, distribute or copy this email. Please delete this email 
from your system and notify the sender immediately by replying to this email.  
If you are not the intended recipient you are notified that disclosing, 
copying, distributing or taking any action in reliance on the contents of this 
information is strictly prohibited.

The Alternative Investments division of UMB Fund Services provides a full range 
of services to hedge funds, funds of funds and private equity funds.  Any tax 
advice in this communication is not intended to be used, and cannot be used, by 
a client or any other person or entity for the purpose of (a) avoiding 
penalties that may be imposed on any taxpayer or (b) promoting, marketing, or 
recommending to another party any matter addressed herein.
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jim Crate via 4D_Tech]

There have been enough reports that they know it happens on both macOS and 
Windows. Multiple reports indicate the processes can continue when RESUME 
PROCESS is called. In this case RESUME PROCESS would have worked, because when 
quitting the server, the processes started again because the quitting process 
resumes them. However, this bug will never be reproducible on demand. The only 
way this bug can be fixed is by 4D engineers looking into the code to figure 
out why sometimes processes delayed with DELAY PROCESS don’t wake up. None of 
us know what kind of mechanism or data structures they are using to track when 
delayed processes should wake up.

Sometimes, as programmers, we just have to take the customer reports of 
something happening and look through our code to see what can possibly trigger 
the condition. I have done that before and discovered the cause of 
non-reproducible bugs. I have also looked and not found code that could cause a 
condition the customer reported.

Jim Crate


> On Jun 12, 2018, at 1:27 PM, Dennis, Neil via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
>> DELAY PROCESS never comes back
> 
> Be sure to submit a bug with 4D when this happens. I submitted one and they 
> said if it isn't reproducible they "can't" fix it. If there is enough 
> complaining about it, I hope they would spend time digging into it and 
> realize it is a problem even if it can't be reproduced on demand.
> 
> Neil
> 
> 
> 
> 
> 
> 
> --
> 
> Privacy Disclaimer: This message contains confidential information and is 
> intended only for the named addressee. If you are not the named addressee you 
> should not disseminate, distribute or copy this email. Please delete this 
> email from your system and notify the sender immediately by replying to this 
> email.  If you are not the intended recipient you are notified that 
> disclosing, copying, distributing or taking any action in reliance on the 
> contents of this information is strictly prohibited.
> 
> The Alternative Investments division of UMB Fund Services provides a full 
> range of services to hedge funds, funds of funds and private equity funds.  
> Any tax advice in this communication is not intended to be used, and cannot 
> be used, by a client or any other person or entity for the purpose of (a) 
> avoiding penalties that may be imposed on any taxpayer or (b) promoting, 
> marketing, or recommending to another party any matter addressed herein.
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Cannon Smith via 4D_Tech]

We just released a new version of our software a week and a half ago. In this 
version I added a palette window on the server which runs On Timer. I also 
added code to most every background process which updates some information 
every time it wakes up from a Delay Process call. The palette window checks 
this information and, if it sees a process that has been asleep longer than one 
minute, resumes the process and sends us a Slack message identifying the server 
and process that was asleep. If the process does not get resumed, another Slack 
message is sent so we can handle the issue manually before our customer knows.

Interestingly, this mechanism was triggered on Sunday morning at 1:01 AM for 9 
servers. All 9 reported one or more stalled processes within a 60 second time 
period.

--
Cannon.Smith
Synergy Farm Solutions Inc.
Hill Spring, AB Canada
403-626-3236




> On Jun 12, 2018, at 1:43 PM, Jim Crate via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> Our server (macOS 10.11, 4D 16.3 HF3 64-bit) hit this bug again last week 
> after being up for about 22 days. There are 2 separate processes that run 
> periodically, both at the same time stopped waking from DELAY PROCESS. The 
> Server window also was not showing the 4D username in the user list, although 
> IP, machine name, and login date/time were displayed.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Keisuke Miyako via 4D_Tech]

22 days does sound suspicious; as if (no proof),
the scheduler is testing if milliseconds "minus" process up-time is "greater 
than" delay time...
such logic would break when milliseconds expressed in signed 32-bit long 
integer flips to negative, after 24 days.
(milliseconds is counted since system boot, not 4D launch)

> 2018/06/13 4:43、Jim Crate via 4D_Tech <4d_tech@lists.4d.com> のメール：
> Our server (macOS 10.11, 4D 16.3 HF3 64-bit) hit this bug again last week 
> after being up for about 22 days.




**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: Delay process bug

[Dennis, Neil via 4D_Tech]

> DELAY PROCESS never comes back

Be sure to submit a bug with 4D when this happens. I submitted one and they 
said if it isn't reproducible they "can't" fix it. If there is enough 
complaining about it, I hope they would spend time digging into it and realize 
it is a problem even if it can't be reproduced on demand.

Neil






--

Privacy Disclaimer: This message contains confidential information and is 
intended only for the named addressee. If you are not the named addressee you 
should not disseminate, distribute or copy this email. Please delete this email 
from your system and notify the sender immediately by replying to this email.  
If you are not the intended recipient you are notified that disclosing, 
copying, distributing or taking any action in reliance on the contents of this 
information is strictly prohibited.

The Alternative Investments division of UMB Fund Services provides a full range 
of services to hedge funds, funds of funds and private equity funds.  Any tax 
advice in this communication is not intended to be used, and cannot be used, by 
a client or any other person or entity for the purpose of (a) avoiding 
penalties that may be imposed on any taxpayer or (b) promoting, marketing, or 
recommending to another party any matter addressed herein.
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Jim Crate via 4D_Tech]

Our server (macOS 10.11, 4D 16.3 HF3 64-bit) hit this bug again last week after 
being up for about 22 days. There are 2 separate processes that run 
periodically, both at the same time stopped waking from DELAY PROCESS. The 
Server window also was not showing the 4D username in the user list, although 
IP, machine name, and login date/time were displayed.

A watchdog process would have to run continuously to avoid being delayed at the 
time when delay timers die. Sucking up that much cpu just to try to be notified 
of a bug that occurs every few weeks doesn’t seem like a great solution. Also, 
because these methods use plugins and probably Begin/End SQL blocks, I can’t 
run them preemptively to see if preemptive threads have the same problem with 
DELAY PROCESS.

Jim Crate


> On Mar 31, 2018, at 8:31 AM, Jeffrey Kain via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> So, the Delay Process bug hit two servers within an hour of each other this 
> morning, after months of not having the issue.  It was weird that both of 
> them hit the bug this morning within a short period of time of each other.
> 
> This was on 16.3 HF2, so the latest and greatest...
> **
> 4D Internet Users Group (4D iNUG)
> FAQ:  http://lists.4d.com/faqnug.html
> Archive:  http://lists.4d.com/archives.html
> Options: https://lists.4d.com/mailman/options/4d_tech
> Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
> **

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: Delay process bug

[Keith Goebel via 4D_Tech]

> On 1/04/2018, at 7:00 am, Jeffrey Kain wrote:
> So, the Delay Process bug hit two servers within an hour of each other this 
> morning, after months of not having the issue.  It was weird that both of 
> them hit the bug this morning within a short period of time of each other.
> This was on 16.3 HF2, so the latest and greatest...

Hi all - this is just an FYI…
Using 4D v16.3 on Win, we have confirmed that a particular stored procedure 
(background tasks) running on “Client” will freeze whereas if it is run on the 
server it does not. 
I’ve been away for a short while but hope to work with 4D soon to try to narrow 
down the cause.
Cheers, Keith
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: https://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[Bernd Fröhlich via 4D_Tech]

John DeSoi:

> Are you having this issue with the server, client, or both? I'm just seeing 
> it with 4D Client. My new theory is that during a long delay 4D Server 
> disconnects the client because there is no activity and 4D Server timeout 
> (selector 13 for SET DATABASE PARAMETER) is reached. This may be wrong 
> because I ran experiments yesterday and still could not reproduce the 
> problem. But I changed the database setting in the dialog from 15 minutes to 
> unlimited and the client did not hang last night as it has the last several. 

I had the problem on the Server (Mac). About once a month my Cron-process would 
delay and never wake up again.
It can´t be an uptime-issue because the 4D-server is restarted every night.
It happend on every time of day, so no consistancy there either.
No idea what the cause may be.

Greetings from Germany,
Bernd Fröhlich

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[John DeSoi via 4D_Tech]

Bernd,

Are you having this issue with the server, client, or both? I'm just seeing it 
with 4D Client. My new theory is that during a long delay 4D Server disconnects 
the client because there is no activity and 4D Server timeout (selector 13 for 
SET DATABASE PARAMETER) is reached. This may be wrong because I ran experiments 
yesterday and still could not reproduce the problem. But I changed the database 
setting in the dialog from 15 minutes to unlimited and the client did not hang 
last night as it has the last several. 

I'm going to test setting this parameter locally for the execution process and 
see that happens.

John DeSoi, Ph.D.


> On Jan 8, 2018, at 1:39 AM, Bernd Fröhlich via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
>> That's not really a solution though, we'd have to write like a process 
>> watching that would have to make a call that process needed to be revived.
> 
> Just make sure the watching process does not depend on using Delay Process 
> for itself.
> 
> Since Delay Process sometimes just behaves like Pause Process, I decided to 
> use Pause Process instead and wake it from another process that uses a window 
> with a timer to do a Resume Process.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: DELAY PROCESS bug

[Bernd Fröhlich via 4D_Tech]

Drew Waddell:

> That's not really a solution though, we'd have to write like a process 
> watching that would have to make a call that process needed to be revived.

Just make sure the watching process does not depend on using Delay Process for 
itself.

Since Delay Process sometimes just behaves like Pause Process, I decided to use 
Pause Process instead and wake it from another process that uses a window with 
a timer to do a Resume Process.

Works fine so far.

Greetings from Germany,
Bernd Fröhlich
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[John DeSoi via 4D_Tech]

After this discussion I ran a test server for several weeks before upgrading 
from 15.4 to 16.3 (Mac 64 bit). No problem running jobs on the server 
interleaved with DELAY PROCESS. After upgrading the production database to 
16.3, a client running this same procedure regularly locks up. It seems to be 
related to backup. The client is frozen and appears to be no longer connected 
to the server (no process in the admin window). The server no longer sees it in 
GET REGISTERED CLIENTS.

I'm going to try adjusting this procedure to ensure it does nothing during 
backup and see if that fixes it. 

John DeSoi, Ph.D.



> On Jan 5, 2018, at 12:33 PM, Cannon Smith via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> I wonder if this might be related to the Server Process Frozen issue lots of 
> us have been experiencing. It sounds like the result is pretty much the same.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: DELAY PROCESS bug

[Drew Waddell via 4D_Tech]

@Cannon It's like Inception... I don't want to go down that path

-Original Message-
From: 4D_Tech [mailto:4d_tech-boun...@lists.4d.com] On Behalf Of Cannon Smith 
via 4D_Tech
Sent: Friday, January 05, 2018 1:44 PM
To: 4D iNug Technical
Cc: Cannon Smith
Subject: Re: DELAY PROCESS bug

> That's not really a solution though, we'd have to write like a process 
> watching that would have to make a call that process needed to be revived.

And then another process to watch the watch dog process in case it stops. And 
then another one to watch the one watching the watch dog. Etc. :-)

--
Cannon.Smith
Synergy Farm Solutions Inc.
Hill Spring, AB Canada
403-626-3236
<can...@synergyfarmsolutions.com>




**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[Jeffrey Kain via 4D_Tech]

Yeah, that's a pretty significant bug. Hopefully they fix it!

I tried it on a Windows Server 2012R2 machine running 4D Remote, and did a 
ctrl-S over the DELAY PROCESS line and then minimized -- same thing.  RESUME 
PROCESS did nothing.


> On Jan 5, 2018, at 1:40 PM, Drew Waddell  wrote:
> 
> Hey Jeff,
> 
> After I typed it up I realized I probably should have said we are working on 
> Windows but also thought the F4 might have been a giveaway.  I was curious 
> about whether it would be different on a Mac.  I tried from another process 
> RESUME PROCESS, BRING TO FRONT, and DELAY PROCESS with passing 0 which seems 
> like it's the equivalent to RESUME PROCESS based on the documentation notes.  
> No luck.
> 
> Ah, here we go, calling SHOW PROCESS brought it back to life.  What's 
> interesting here is that when calling PROCESS PROPERTIES it said the process 
> was visible.
> 
> That's not really a solution though, we'd have to write like a process 
> watching that would have to make a call that process needed to be revived.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[Cannon Smith via 4D_Tech]

> That's not really a solution though, we'd have to write like a process 
> watching that would have to make a call that process needed to be revived.

And then another process to watch the watch dog process in case it stops. And 
then another one to watch the one watching the watch dog. Etc. :-)

--
Cannon.Smith
Synergy Farm Solutions Inc.
Hill Spring, AB Canada
403-626-3236





**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: DELAY PROCESS bug

[Randy Engle via 4D_Tech]

I might chime in here as well:

"Sometimes", under certain conditions, HIDE PROCESS behaves the same as DELAY 
PROCESS
e.g. code should keep executing after HIDE PROCESS, but it gets "delayed" 
instead.

I've not narrowed in down but we are certainly seeing this happen.

Are these issues related?   Hmmm... maybe/maybe not.

Randy Engle, Director
XC2 Software LLC – XC2LIVE!



**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

RE: DELAY PROCESS bug

[Drew Waddell via 4D_Tech]

Hey Jeff,

After I typed it up I realized I probably should have said we are working on 
Windows but also thought the F4 might have been a giveaway.  I was curious 
about whether it would be different on a Mac.  I tried from another process 
RESUME PROCESS, BRING TO FRONT, and DELAY PROCESS with passing 0 which seems 
like it's the equivalent to RESUME PROCESS based on the documentation notes.  
No luck.

Ah, here we go, calling SHOW PROCESS brought it back to life.  What's 
interesting here is that when calling PROCESS PROPERTIES it said the process 
was visible.

That's not really a solution though, we'd have to write like a process watching 
that would have to make a call that process needed to be revived.

Drew

-Original Message-
From: 4D_Tech [mailto:4d_tech-boun...@lists.4d.com] On Behalf Of Jeffrey Kain 
via 4D_Tech
Sent: Friday, January 05, 2018 1:34 PM
To: 4D iNug Technical
Cc: Jeff Kain
Subject: Re: DELAY PROCESS bug

Hmmm... another issue with DELAY PROCESS.

We have the issue in v16 where stored procedures that are delayed will 
sometimes never wake up. In our case, it's our mirror process. Sometimes the 
stored procedure just stops, forever delayed.

It is possible to resume it by opening the Runtime Explorer and typing in a 
RESUME PROCESS command.

Just tried your example on Mac v16 32-bit and didn't have the problem (I hid 4D 
since there's no real minimize). Our stored procedure issues are all on 
Windows, however.

Jeff

--
Jeffrey Kain
jeffrey.k...@gmail.com




> On Jan 5, 2018, at 1:24 PM, Drew Waddell via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> When the application is minimized wait 30 seconds or so, long enough to know 
> the DELAY PROCESS should have returned.  Maximize the application and notice 
> that you have not hit your TRACE. If you hit Ctrl+Shift+Right Click you'll 
> see a P_X method that you are unable to break into. If you check the Process 
> state, it is 2.  We have not found a way to get the process to come back to 
> life.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**
**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[Jeffrey Kain via 4D_Tech]

Hmmm... another issue with DELAY PROCESS.

We have the issue in v16 where stored procedures that are delayed will 
sometimes never wake up. In our case, it's our mirror process. Sometimes the 
stored procedure just stops, forever delayed.

It is possible to resume it by opening the Runtime Explorer and typing in a 
RESUME PROCESS command.

Just tried your example on Mac v16 32-bit and didn't have the problem (I hid 4D 
since there's no real minimize). Our stored procedure issues are all on 
Windows, however.

Jeff

--
Jeffrey Kain
jeffrey.k...@gmail.com




> On Jan 5, 2018, at 1:24 PM, Drew Waddell via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> When the application is minimized wait 30 seconds or so, long enough to know 
> the DELAY PROCESS should have returned.  Maximize the application and notice 
> that you have not hit your TRACE. If you hit Ctrl+Shift+Right Click you'll 
> see a P_X method that you are unable to break into. If you check the Process 
> state, it is 2.  We have not found a way to get the process to come back to 
> life.

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**

Re: DELAY PROCESS bug

[Cannon Smith via 4D_Tech]

Hi Drew,

I wonder if this might be related to the Server Process Frozen issue lots of us 
have been experiencing. It sounds like the result is pretty much the same.

--
Cannon.Smith
Synergy Farm Solutions Inc.
Hill Spring, AB Canada
403-626-3236




> On Jan 5, 2018, at 11:24 AM, Drew Waddell via 4D_Tech <4d_tech@lists.4d.com> 
> wrote:
> 
> We are putting in a ticket for this but wanted to see if anyone else has 
> encountered this and found a work around.  We haven't investigated it too 
> much to determine if it's an issue outside of interpreted mode but it has us 
> a little worried.
> 
> Put this code snippet in a method and hit the green arrow button.
> 
> TRACE // Press F4 and minimize
> DELAY PROCESS(Current process;600)
> TRACE
> 
> When the application is minimized wait 30 seconds or so, long enough to know 
> the DELAY PROCESS should have returned.  Maximize the application and notice 
> that you have not hit your TRACE.  If you hit Ctrl+Shift+Right Click you'll 
> see a P_X method that you are unable to break into.  If you check the Process 
> state, it is 2.  We have not found a way to get the process to come back to 
> life.
> 
> We have reproduced this in v15 and v16.
> 

**
4D Internet Users Group (4D iNUG)
FAQ:  http://lists.4d.com/faqnug.html
Archive:  http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub:  mailto:4d_tech-unsubscr...@lists.4d.com
**