We met the case of that value of "entry->state" in function, dns_check_entry(u8_t i) was 58 or 59.

[bini2007]

We met the case of that value of "entry->state" in function, dns_check_entry(u8_t i) was 58 or 59.

Our test env.:

• hw: ESP32,
• sw: esp-idf ,
• condition: use SNTP and test with/without internet connection.

< Problem Description >
When we tested our target without no antenna (it seems to be without wifi connection.),
the following happened.

So after we debugged with JTAG debugger, we found the following debugged fact.

The debugged fact is as followings:

I met the case of that value of "entry->state"
in dns_check_entry(u8_t i) function in the source, dns.c was 58 or 59.

So our program went to the default in case statement and met the LWIP_ASSERT() and fianally paniced.

In the end, we didn't know why
but first of all, we added the following codes in the case of the value, 58 and 59 in the function, dns_check_entry(u8_t i).

case 58:
case 59:
{
  LWIP_DEBUGF(DNS_DEBUG, ("dns_check_entry: \"%s\": flush\n", entry->name));
  /* flush this entry, there cannot be any related pending entries in this state */
  entry->state = DNS_STATE_UNUSED;
}
break;

As you know, we added the same code as the case, DNS_STATE_DONE in the switch statement.
Then the above refered problem didn't happened.

But we still don't know why the case of the value 58 or 59 happend when without anntenna (wifi connection)
so anyone, please explain to us about this problem and give us more solutions if exist.