UDP Query keeps timing out

[krishnak]

            while i<10:
                try:
                    client = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
                    client.settimeout(5)
                    client.sendto(query,addr)
                    print("data sent")                 
                    data = client.recv(48)
                    client.close()
                    print("data received")

I keep getting time out thrown, the code used to work over night can anyone suggest a fix

data sent
[Errno 110] ETIMEDOUT
data sent
[Errno 110] ETIMEDOUT
data sent

[krishnak]

It appears that a fresh query is successful only on the fourth attempt and there are things I may have to do to get it working inside async io

[robert-hh]

Please tell, which board & MicroPython version you use. Besides that: Are you sure that the server responds?

[krishnak]

Thank you I am running micro python ver 1.19 on Weimos ESp8266, the server is pool.ntp.org or its variant domains like 0.uk.pool.ntp.org, I have tested it in stand alone mode i.e no async nothing just standalone file to avoid confusion with my other application code. The stand alone function code is as below. The server responds - please see log below, the IP address returned by DNS is the same from the second try till it succeeds on the 8th try

def configuretime(tzoffset=0):
    NTP_QUERY = bytearray(48)
    NTP_QUERY[0] = 0x1b
    addr = socket.getaddrinfo(host, 123)[0][-1]
    s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
    s.settimeout(1)
    res = s.sendto(NTP_QUERY, addr)
    msg = s.recv(48)
    s.close()
    t = struct.unpack("!I", msg[40:44])[0]
    t -= referencetime
    t+=tzoffset
    tm = utime.localtime(t) 
    tm = tm[0:3] + (0,) + tm[3:6] + (0,)
    machine.RTC().datetime(tm)

 NetWorkWrapper  <DEBUG> Connected to router 
('178.215.228.24', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
[Errno 110] ETIMEDOUT
('193.150.34.2', 123)
(2022, 9, 18, 8, 49, 36, 6, 261)

NetWorkWrapper  <DEBUG> Connected to router 
('192.46.210.39', 123)
[Errno 110] ETIMEDOUT
('162.159.200.123', 123)
[Errno 110] ETIMEDOUT
('162.159.200.123', 123)
[Errno 110] ETIMEDOUT
('162.159.200.123', 123)
[Errno 110] ETIMEDOUT
('162.159.200.123', 123)
(2022, 9, 18, 8, 53, 53, 6, 261)

[davefes]

I had more success with a 10 second timeout.

I also put the core of code similar to yours in a while loop that would bail-out after 5 attempts. I still occasionally get a fail.

[krishnak]

With 10 second time out WDT kicks in to restart the device.

The following is NON blocking socket i.e time out set to 0

NetWorkWrapper  <DEBUG> Connected to router 
('129.154.46.154', 123)
[Errno 11] EAGAIN
('162.159.200.123', 123)
[Errno 11] EAGAIN
('152.70.69.232', 123)
[Errno 11] EAGAIN
('162.159.200.123', 123)
[Errno 11] EAGAIN
('162.159.200.1', 123)
[Errno 11] EAGAIN
('162.159.200.1', 123)
[Errno 11] EAGAIN
('178.215.228.24', 123)
[Errno 11] EAGAIN
('162.159.200.1', 123)
[Errno 11] EAGAIN
('162.159.200.1', 123)
[Errno 11] EAGAIN
('178.215.228.24', 123)
[Errno 11] EAGAIN
(2022, 9, 18, 14, 51, 35, 6, 261)
Main  <DEBUG> 15920

[krishnak]

Ok, I vaguely remember reading in the docs that asyncio using UDP is flaky or not implemented and has to be handled with polling.

The following code works every time on the first attempt

def configuretime(tzoffset=0):
    NTP_QUERY = bytearray(48)
    NTP_QUERY[0] = 0x1b
    poller = select.poll()

    addr = socket.getaddrinfo(host, 123)[0][-1]
    print(addr)
    s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
    s.setblocking(False)
    poller.register(s, select.POLLIN)
#    s.settimeout(0)

    res = s.sendto(NTP_QUERY, addr)
    for response in poller.ipoll(300):
        msg = s.recv(48)
        s.close()
        t = struct.unpack("!I", msg[40:44])[0]
        t -= referencetime
        t+=tzoffset
        tm = utime.localtime(t) 
        tm = tm[0:3] + (0,) + tm[3:6] + (0,)
        machine.RTC().datetime(tm)
        break

[davefes]

every time on the first attempt
I probably get one failure after 200-300 tries, I'll bookmark your code

[krishnak]

I spoke too soon, the code is not setting the time, when you connect via rshell it sets local time, the for loop doesn't even get executed

[krishnak]

The polling method also requires 5 data packets to return a result. The only difference now is that there is no Exception being thrown, the socket after polling silently proceeds to the next statement which is break :)

[krishnak]

Just an update the above code which I have posted with polling works as intended on the first attempt, with a minor modification as below

async def RequestTimefromNtp(host='0.de.pool.ntp.org'):
        syncinterval = const(12*60*60)
        referencetime = const(3155673600)  # Reference time
        while True:
            try:
                NTP_QUERY = bytearray(48)
                NTP_QUERY[0] = 0x1b

                for i in range(10):
                    addr = socket.getaddrinfo(host, 123)[0][-1]
                    s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
                    poller = select.poll()
                    poller.register(s,select.POLLIN)
                    print(addr)
                    s.sendto(NTP_QUERY, addr)
                    
                    if poller.poll(1000):  
                        msg = s.recv(48)
                        s.close()
                        t = struct.unpack("!I", msg[40:44])[0]
                        t -= referencetime
                        t+=tzoffset
                        tm = utime.localtime(t) 
                        tm = tm[0:3] + (0,) + tm[3:6] + (0,)
                        machine.RTC().datetime(tm)
                        Logger.Log(tag," Time Synchronised ",Logger.INFO)
                        break         

            except Exception as e:
                Logger.Log(tag,"Error setting system time "+str(e),Logger.ERROR)
                await asyncio.sleep(1)
                continue
            await asyncio.sleep(syncinterval)