Anton Zverev: I'm about cloud boards as a buyer of resources.
Why not? While uploading history - the processor is also working and time is spent.
Hopefully, to save traffic and increase speed, your job will be given to the agent who already has your broker's history uploaded by previous jobs. I hope MQ have done that...
Why not? While uploading history - the processor is also working and time is spent.
Hopefully, to save traffic and increase speed, your assignment will be given to the agent who has your broker's history already uploaded by previous assignments. Hopefully MQ have done this...
All agents have logs cut off with entries like this:
DR 008:43:31.562 Network connected to 1.agents.mql5.com
LJ 008:43:39.097 Network connected to 2.agents.mql5.com
NE 008:43:47.986 Network connected to 4.agents.mql5.com
DL 008:43:55.416 Network connected to 5.agents.mql5.com
CJ 008:44:04.109 Network MQL5 Cloud Network server 4.agents.mql5.com selected after rescan (ping 20 ms)
Although (if you look at old logs) there should be two more lines like this:
LM 002:11:25.047 Network connected to 4.agents.mql5.com
MI 002:11:25.092 Network authorized on 4.agents.mql5.com for barabashkakvn
Isn't this the reason for the lack of assignments (no assignments at all) since about 1 April?
It used to be that after connecting to the network, the agents would ping the server every 30 seconds - informing the server something like "agent # so-and-so is alive". Now there is no traffic after connecting to the cloud. That is, yesterday agents were created at about 21:57 and after connecting to the network all agents had 4/1 Kb traffic. And here's what's happening today:
Could this be why the agents are not getting assignments? Meanwhile, all night long, the computer
2016.04.2422:10:14.592 Terminal MetaTrader 5 x64 build 1325 started (MetaQuotes Software Corp.)
2016.04.2422:10:14.593 Terminal Windows 10 Pro (x64 based PC), IE 13.12, UAC, Intel Core i3-3120 M @ 2.50 GHz, RAM: 911 / 3981 Mb, HDD: 45531 / 233310 Mb, GMT+02:002016.04.2422:10:14.593 Terminal C:\Users\KVN\AppData\Roaming\MetaQuotes\Terminal\0 CFEFA8410765D70FC53545BFEFB44F4
Was running with about 1.4 GB of free RAM and was plugged in on charge all the time. Here's one of the agent logs:
OM 021:45:54.999 Startup access rights to common directory successfully checked
IF 021:45:55.001 Startup service start initialized
NQ 021:45:55.002 Startup Create startup thread
QS 021:45:55.002 Startup thread successfully created
QK 021:45:55.002 Startup MetaTester 5 x64 build 1325 (22 Apr 2016)
FN 021:45:55.010 Server MetaTester 5 started on 0.0.0.0:2000
EK 021:45:55.014 Startup initialization finished
RQ 021:45:55.015 Startup startup thread finished successfully
DF 021:45:56.692 Network connected to 1.agents.mql5.com
MN 021:46:05.149 Network connected to 2.agents.mql5.com
EI 021:46:14.071 Network connected to 4.agents.mql5.com
MP 021:46:21.500 Network connected to 5.agents.mql5.com
QK 021:46:30.302 Network connected to 1.agents.mql5.com
NS 021:46:37.837 Network connected to 2.agents.mql5.com
DJ 021:46:46.727 Network connected to 4.agents.mql5.com
CE 021:46:54.148 Network connected to 5.agents.mql5.com
IL 021:47:02.889 Network connected to 1.agents.mql5.com
ED 021:47:10.427 Network connected to 2.agents.mql5.com
FO 021:47:19.308 Network connected to 4.agents.mql5.com
LF 021:47:26.734 Network connected to 5.agents.mql5.com
LL 021:47:35.427 Network MQL5 Cloud Network server 4.agents.mql5.com selected after rescan (ping 19 ms)
As you can see, there are no entries at all for today's 24 hours. And in the agent folder, there is a log file for yesterday only:
OM 021:45:54.999 Startup access rights to common directory successfully checked
IF 021:45:55.001 Startup service start initialized
NQ 021:45:55.002 Startup Create startup thread
QS 021:45:55.002 Startup thread successfully created
QK 021:45:55.002 Startup MetaTester 5 x64 build 1325 (22 Apr 2016)
FN 021:45:55.010 Server MetaTester 5 started on 0.0.0.0:2000
EK 021:45:55.014 Startup initialization finished
RQ 021:45:55.015 Startup startup thread finished successfully
DF 021:45:56.692 Network connected to 1.agents.mql5.com
MN 021:46:05.149 Network connected to 2.agents.mql5.com
EI 021:46:14.071 Network connected to 4.agents.mql5.com
MP 021:46:21.500 Network connected to 5.agents.mql5.com
QK 021:46:30.302 Network connected to 1.agents.mql5.com
NS 021:46:37.837 Network connected to 2.agents.mql5.com
DJ 021:46:46.727 Network connected to 4.agents.mql5.com
CE 021:46:54.148 Network connected to 5.agents.mql5.com
IL 021:47:02.889 Network connected to 1.agents.mql5.com
ED 021:47:10.427 Network connected to 2.agents.mql5.com
FO 021:47:19.308 Network connected to 4.agents.mql5.com
LF 021:47:26.734 Network connected to 5.agents.mql5.com
LL 021:47:35.427 Network MQL5 Cloud Network server 4.agents.mql5.com selected after rescan (ping 19 ms)
As you can see, there are no entries at all for today's 24 hours. And this file is from yesterday. In general, it seems that after connecting to the cloud, the cloud completely closes access to agents.
Judging by the log, the agent has no problems. It found the nearest server, connected and is waiting for the task.
It used to be that agents used to ping the server every 30 seconds after connecting to the network
That's not true anymore, and for quite a long time. Now agents do not break connection with the selected server and are in constant readiness to receive the task.
As you can see, there's no record of that at all in the last 24 hours. And in the agent's folder, there is a log file for yesterday only:
The log is written to disc as sparingly as possible, especially when the agent is completely inactive. You can restart the agent service and the log will be dumped to disc.
I'm about cloud boards as a buyer of resources.
Why not? While uploading history - the processor is also working and time is spent.
Hopefully, to save traffic and increase speed, your job will be given to the agent who already has your broker's history uploaded by previous jobs. I hope MQ have done that...
Why not? While uploading history - the processor is also working and time is spent.
Hopefully, to save traffic and increase speed, your assignment will be given to the agent who has your broker's history already uploaded by previous assignments. Hopefully MQ have done this...
All agents have logs cut off with entries like this:
Although (if you look at old logs) there should be two more lines like this:
Isn't this the reason for the lack of assignments (no assignments at all) since about 1 April?
It used to be that after connecting to the network, the agents would ping the server every 30 seconds - informing the server something like "agent # so-and-so is alive". Now there is no traffic after connecting to the cloud. That is, yesterday agents were created at about 21:57 and after connecting to the network all agents had 4/1 Kb traffic. And here's what's happening today:
Could this be why the agents are not getting assignments? Meanwhile, all night long, the computer
Was running with about 1.4 GB of free RAM and was plugged in on charge all the time. Here's one of the agent logs:
As you can see, there are no entries at all for today's 24 hours. And in the agent folder, there is a log file for yesterday only:
All agents' logs are terminated with such entries:
Although (if you look at old logs) there should be two more lines like this:
Isn't this the reason for the lack of assignments (no assignments at all) since about 1 April?
Attach the agent log file.
I'm attaching it:
As you can see, there are no entries at all for today's 24 hours. And this file is from yesterday. In general, it seems that after connecting to the cloud, the cloud completely closes access to agents.Attached:
Judging by the log, the agent has no problems. It found the nearest server, connected and is waiting for the task.
That's not true anymore, and for quite a long time. Now agents do not break connection with the selected server and are in constant readiness to receive the task.
The log is written to disc as sparingly as possible, especially when the agent is completely inactive. You can restart the agent service and the log will be dumped to disc.
Even MetaTrader 5 Strategy Tester has updated itself to build 1304, and since 18 April agents have only zeros in the "PR" column:
Pardon for off-topic, but what was the maximum PR of your Phenom II X6 1075T? I was looking at it at one time just for this purpose.
is the mql5 cloud service even working?
2 weeks of idle CPU without load. Only 12 tests passed.
Everything is ok in the logs. Or is there just so little demand for cloud computing?
I assume that it would not hurt to make it possible to minimise the agent manager to the tray, well, we can probably do it ourselves at startup )).
GPUs don't seem to be connected, so we'll test only on regular processors.