Unable to look up TXT record for host ****.****.mongodb.net from GCP ap

Pranav_Jariwala · 2022-06-11T12:40:22.070Z

I have a single node GCP cluster with my spring-boot app running. I recently created a new DB [M10] cloud atlas and trying to connect it via my app. However, I keep getting the following error - screenshot attached

Screenshot 2022-06-11 at 14.27.531920×961 289 KB

However, I am able to connect to my old DBs [M20 and M10] via the same GCP cluster and same spring-boot app.

I am unable to figure out what could be the reason? Any idea?

Thanks in advance
Pranav

Jeffrey_Yemin · 2022-06-11T17:26:30.208Z

Is there a final “Caused by” that’s cut off from that screen shot? That could help diagnose the issue. It’s possible that you’re running into https://jira.mongodb.org/browse/JAVA-4018, in which case updating the Java driver to 4.5.1 or later will fix it.

Is this a self-managed cluster, or are using Atlas, or some other service? I am pretty sure that Atlas always registers a TXT record in DNS, but other services may not.

Pranav_Jariwala · 2022-06-11T20:43:32.721Z

final caused by :

Screenshot 2022-06-11 at 22.39.332870×1014 414 KB

I am using Atlas and spring-boot-starter-data-mongodb 4.6.0

the other 2 DBs are also from Atlas and I am able to connect to them

Pranav_Jariwala · 2022-06-11T20:45:15.338Z

from the bug you referenced it looks like it may have re-appeared in 4.6.0.

Pranav_Jariwala · 2022-06-11T21:08:53.587Z

Pranav_Jariwala:

4.6.0

sorry i am using spring-boot-starter-data-mongodb 4.7.0 which internally uses 4.6.0 mongo java driver

Jeffrey_Yemin · 2022-06-11T21:14:07.081Z

You can see from the stack trace that it’s using mongodb-driver-core-4.4.0.jar.

Pranav_Jariwala · 2022-06-12T14:27:21.825Z

Thanks for spotting that. But after upgrading I am still getting DNS issues

Screenshot 2022-06-12 at 16.26.112672×864 316 KB

steevej · 2022-06-12T14:34:13.474Z

It is not the same address as before.

Pranav_Jariwala · 2022-06-12T14:36:31.544Z

yeah… i deleted the previous DB and created new one

Pranav_Jariwala · 2022-06-12T15:28:25.000Z

any idea what might be wrong? I am using the same old username/pass combination that I have for the other 2 DB [no special character] . didn’t find anything relevant online

Pranav_Jariwala · 2022-06-12T15:39:46.625Z

There’s something wrong with the new Atlas DB instances. There seems to be an issue only while connecting with the new DB. The old DB connection are working perfectly fine

Jeffrey_Yemin · 2022-06-12T16:10:38.663Z

Notice that now the TXT record is no longer failing (that happens first) but SRV lookup is still failing. Using dig, I’m able to resolve both TXT and SRV records for this host:

~$ dig SRV _mongodb._tcp.hk-prod.h8dkbpg.mongodb.net

; <<>> DiG 9.10.6 <<>> SRV _mongodb._tcp.hk-prod.h8dkbpg.mongodb.net
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 19870
;; flags: qr rd ra; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 512
;; QUESTION SECTION:
;_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. IN SRV

;; ANSWER SECTION:
_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. 60 IN SRV 0 0 27017 ac-gokumri-shard-00-00.h8dkbpg.mongodb.net.
_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. 60 IN SRV 0 0 27017 ac-gokumri-shard-00-01.h8dkbpg.mongodb.net.
_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. 60 IN SRV 0 0 27017 ac-gokumri-shard-00-02.h8dkbpg.mongodb.net.

;; Query time: 167 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Sun Jun 12 12:08:08 EDT 2022
;; MSG SIZE  rcvd: 256

~$ dig TXT  _mongodb._tcp.hk-prod.h8dkbpg.mongodb.net

; <<>> DiG 9.10.6 <<>> TXT _mongodb._tcp.hk-prod.h8dkbpg.mongodb.net
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 7702
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 512
;; QUESTION SECTION:
;_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. IN TXT

;; ANSWER SECTION:
_mongodb._tcp.hk-prod.h8dkbpg.mongodb.net. 60 IN TXT "authSource=admin&replicaSet=atlas-tbk5qg-shard-0"

;; Query time: 35 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Sun Jun 12 12:08:57 EDT 2022
;; MSG SIZE  rcvd: 131

Can you try the same from the same server on which you’re running the Java application?

Pranav_Jariwala · 2022-06-12T16:47:48.407Z

looks like dig worked.

Screenshot 2022-06-12 at 18.45.311752×1494 139 KB

then maybe some issue with com.mongodb.internal.dns.DefaultDnsResolver ?

Jeffrey_Yemin · 2022-06-12T17:26:16.214Z

I think it is something in your environment. I am able to connect successfully to your cluster from both the mongo shell and a Java application.

Pranav_Jariwala · 2022-06-12T18:14:22.925Z

If the issue was with the environment wouldn’t there be an issue connecting to other Atlas DB? [as mentioned previously, I am able to connect to the other 2 DB via the same GCP server, same application]

Jeffrey_Yemin · 2022-06-12T19:08:12.138Z

Not necessarily, but I don’t have a hypothesis that explains what both of us are seeing. I think at this point you should open a support ticket (https://www.mongodb.com/docs/manual/support/) to try to get to the root cause.

Regards,
Jeff

Robert_Timm · 2022-07-22T11:20:16.628Z

Having the same issue as Pranav. GCP is failing to connect, works on my local environment. GCP works when connecting to older instances of Atlas.

org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'appConfig' defined in URL [jar:file:/opt/app/server/app.jar!/BOOT-INF/classes!/com/devnotes/server/app/AppConfig.class]: Bean instantiation via constructor failed; nested exception is org.springframework.beans.BeanInstantiationException: Failed to instantiate [com.devnotes.server.app.AppConfig]: Constructor threw exception; nested exception is com.mongodb.MongoConfigurationException: Unable to look up TXT record for host dev.7f***.mongodb.net at org.springframework.beans.factory.support.ConstructorResolver.instantiate

Robert_Timm · 2022-07-22T11:21:28.009Z

Nevermind, just started happening in my local environment as well.

Dominik_Knoll · 2023-05-16T19:53:37.584Z

I’m facing the same issue, but did not find any solution on the forum here which works.
I can connect to a DB on cluster0.dnqegeu.mongodb.net or cluster0.xfzokj5.mongodb.net with the MongoDB Compass application,
BUT my java spring application fails to connect to a URI with mongodb+srv://
While the mongodb driver should use the SRV entries of the DNS, it looks for a TXT entry.
The library in use is mongodb-driver-core-4.8.2.jar
Any suggestion please?

steevej · 2023-05-17T01:25:23.779Z

The scheme mongodb+srv involves 2 types of DNS records. A TXT record which supplied connection string parameter and SRV records which provides a list of hosts to connect to. Looking for a TXT is thus normal behaviour.

What is the error message?

As for the 2 clusters you share I get:

;QUESTION
cluster0.dnqegeu.mongodb.net. IN ANY
;ANSWER
cluster0.dnqegeu.mongodb.net. 60 IN TXT "authSource=admin&replicaSet=atlas-c64af7-shard-0"
cluster0.dnqegeu.mongodb.net. 60 IN SRV 0 0 27017 ac-eheduyb-shard-00-00.dnqegeu.mongodb.net.
cluster0.dnqegeu.mongodb.net. 60 IN SRV 0 0 27017 ac-eheduyb-shard-00-01.dnqegeu.mongodb.net.
cluster0.dnqegeu.mongodb.net. 60 IN SRV 0 0 27017 ac-eheduyb-shard-00-02.dnqegeu.mongodb.net.

and

;QUESTION
cluster0.xfzokj5.mongodb.net. IN ANY
;ANSWER
cluster0.xfzokj5.mongodb.net. 60 IN TXT "authSource=admin&replicaSet=atlas-kymzsm-shard-0"
cluster0.xfzokj5.mongodb.net. 60 IN SRV 0 0 27017 ac-1b0f5hn-shard-00-00.xfzokj5.mongodb.net.
cluster0.xfzokj5.mongodb.net. 60 IN SRV 0 0 27017 ac-1b0f5hn-shard-00-01.xfzokj5.mongodb.net.
cluster0.xfzokj5.mongodb.net. 60 IN SRV 0 0 27017 ac-1b0f5hn-shard-00-02.xfzokj5.mongodb.net.

So both looks correct.