[Alfresco] After restarting the machine the searches (SOLR) do not work.

[evaristorivi]

Platform

AWS

bndiagnostic ID know more about bndiagnostic ID

032b192a-77d9-7b4b-7508-7aa619a8957f

bndiagnostic output

[Thu Apr 27 18:26:07.282749 2023] [auth_basic:error] [pid 40713:tid
140373632386816] [client **ip_address**:39060] AH01617: user admin:
authentication failure for "/solr/": Password Mismatch

[Thu Apr 27 18:29:34.969747 2023] [proxy:error] [pid 39110:tid 140373741491968]
(111)Connection refused: AH00957: http: attempt to connect to
**ip_address**:8983 (127.0.0.1:8983) failed

bndiagnostic was not useful. Could you please tell us why?

the problem persists.

Describe your issue as much as you can

Searches do not work and this log error appears after rebooting the machine:

        ... 1 more
2023-04-27 19:03:41,803  ERROR [extensions.webscripts.AbstractRuntime] [http-nio-8080-exec-10] Exception from executeScript: 03270003 Wrapped Exception (with status template): 03270047 Failed to execute s
cript 'classpath*:alfresco/templates/webscripts/org/alfresco/slingshot/documentlibrary/doclist.get.js': 03270046 Failed to execute search:  +@cm\:modified:[2023\-4\-20T00\:00\:00.000 TO 2023\-4\-27T23\:59
\:59.999] +@cm\:modifier:"admin" +TYPE:"cm:content" -TYPE:"cm:systemfolder" -TYPE:"fm:forums" -TYPE:"fm:forum" -TYPE:"fm:topic" -TYPE:"fm:post" +(TYPE:"content" OR TYPE:"app:filelink" OR TYPE:"folder")
org.springframework.extensions.webscripts.WebScriptException: 03270003 Wrapped Exception (with status template): 03270047 Failed to execute script 'classpath*:alfresco/templates/webscripts/org/alfresco/sl
ingshot/documentlibrary/doclist.get.js': 03270046 Failed to execute search:  +@cm\:modified:[2023\-4\-20T00\:00\:00.000 TO 2023\-4\-27T23\:59\:59.999] +@cm\:modifier:"admin" +TYPE:"cm:content" -TYPE:"cm:s
ystemfolder" -TYPE:"fm:forums" -TYPE:"fm:forum" -TYPE:"fm:topic" -TYPE:"fm:post" +(TYPE:"content" OR TYPE:"app:filelink" OR TYPE:"folder")
        at org.springframework.extensions.webscripts.AbstractWebScript.createStatusException(AbstractWebScript.java:1139)
        at org.springframework.extensions.webscripts.DeclarativeWebScript.execute(DeclarativeWebScript.java:171)
        at org.alfresco.repo.web.scripts.RepositoryContainer.lambda$transactionedExecute$2(RepositoryContainer.java:553)

[gongomgra]

Hi @evaristorivi,

Thanks for using Bitnami. According to the bndiagnostic output your shared, you are using our Alfresco image 7.3.1-20, but I see we have recently released a couple of extra revisions, and that 7.3.1-22 is the latest image available as you can check in our website. Can you try the latest image in case the issue is solved? If not, can you please give us more details on how to reproduce the issue? Is it only needed to launch a fresh new server and once it has been fully initialized, restarting the machine causes the error?

[evaristorivi]

Hello @gongomgra ,
To reproduce the error, just launch the most current bitnami alfresco image of aws, and once deployed, do a migration from an alfresco running on another machine, just the data and keystore, and reboot the machine.

but I have solved it. I had two problems.
On the one hand, we had a problem with the bitnami script that raises the solr service.
In /opt/bitnami/scripts/libalfresco.sh

is_alfresco_search_services_running() {
    local status
    status=$("${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status)
    local regex
    regex=".*Solr process.*"
    if [[ "$status" =~ $regex ]]; then
        true
    else
        false
    fi
}

${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status gave a random false positive:
Found 1 Solr nodes: Solr process 28130 from /opt/bitnami/alfresco-search-services/solr/bin/solr-8983.pid not found.
and this made bitnami think that the service was running, so after a restart of the machine, it never got solr up.

The fix has been to change the function to:

is_alfresco_search_services_running() {
    local status
    status=$("${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status)
    local regex
    regex=".*Solr process.*"
    real_status=$(ps -ef | grep solr | grep -v grep)
    if [ $? -eq 0 ]; then
        true
    else
        false
    fi
}

This way it always recognises correctly whether it is running or not.
now the start and status scripts inside /opt/bitnami/scripts/alfresco-search-services work correctly.

On the other hand, in order for the communication between solr and alfresco to work correctly, they need to know the secret that is in /opt/bitnami/tomcat/shared/classes/alfresco-global.properties
solr.sharedSecret=*******

But solr didn't know which one was alfresco.
I have edited /opt/bitnami/scripts/alfresco-env.sh
and I have changed
export ALFRESCO_SEARCH_SERVICE_SECRET="${ALFRESCO_SEARCH_SERVICE_SECRET:-bitnami}"
by
export ALFRESCO_SEARCH_SERVICE_SECRET="*******************************"

[gongomgra]

Hi @evaristorivi,

Sorry for the delay on getting back to this. Thank you so much for your investigation and for sharing your solution with us. I'm glad you could make your Alfresco application properly work. I have just filed an internal task for our engineering team to fix the issues you shared. They will keep this ticket updated, and in the meantime I will move it to on-hold status for the stale bot not to close it automatically.

[victorcarabez]

Hello @gongomgra , To reproduce the error, just launch the most current bitnami alfresco image of aws, and once deployed, do a migration from an alfresco running on another machine, just the data and keystore, and reboot the machine.

but I have solved it. I had two problems. On the one hand, we had a problem with the bitnami script that raises the solr service. In /opt/bitnami/scripts/libalfresco.sh

is_alfresco_search_services_running() {
    local status
    status=$("${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status)
    local regex
    regex=".*Solr process.*"
    if [[ "$status" =~ $regex ]]; then
        true
    else
        false
    fi
}

${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status gave a random false positive: Found 1 Solr nodes: Solr process 28130 from /opt/bitnami/alfresco-search-services/solr/bin/solr-8983.pid not found. and this made bitnami think that the service was running, so after a restart of the machine, it never got solr up.

The fix has been to change the function to:

is_alfresco_search_services_running() {
    local status
    status=$("${ALFRESCO_SEARCH_SERVICES_BASE_DIR}/solr/bin/solr" status)
    local regex
    regex=".*Solr process.*"
    real_status=$(ps -ef | grep solr | grep -v grep)
    if [ $? -eq 0 ]; then
        true
    else
        false
    fi
}

This way it always recognises correctly whether it is running or not. now the start and status scripts inside /opt/bitnami/scripts/alfresco-search-services work correctly.

On the other hand, in order for the communication between solr and alfresco to work correctly, they need to know the secret that is in /opt/bitnami/tomcat/shared/classes/alfresco-global.properties solr.sharedSecret=*******

But solr didn't know which one was alfresco. I have edited /opt/bitnami/scripts/alfresco-env.sh and I have changed export ALFRESCO_SEARCH_SERVICE_SECRET="${ALFRESCO_SEARCH_SERVICE_SECRET:-bitnami}" by export ALFRESCO_SEARCH_SERVICE_SECRET="*******************************"

The second workaround described in "On the other hand, in order for the communication between solr and alfresco to work correctly..." doesn't works in a bitnami VM image bitnami-alfresco-7.4.0-r0-debian-11-amd64.ova nor bitnami-alfresco-7.4.0-r1-debian-11-amd64.ova

When check status sudo /opt/bitnami/alfresco-search-services/solr/bin/solr status, it returns:
Found 1 Solr nodes:

Solr process 1761 running on port 8983

ERROR: Failed to get system information from http://localhost:8983/solr due to: org.apache.http.client.ClientProtocolException: Expected JSON response from server but received:
HTTP ERROR 403
Problem accessing /solr/admin/info/system. Reason:
Authentication failure: secret method has been selected, use the right request header with the secret word

The /opt/bitnami/scripts/alfresco-env.sh file has been changed:

# Search Services configuration
export ALFRESCO_SOLR_HOST="${ALFRESCO_SOLR_HOST:-localhost}"
export ALFRESCO_SOLR_PORT="${ALFRESCO_SOLR_PORT:-8983}"
ALFRESCO_SECRET="${ALFRESCO_SECRET:-"${ALFRESCO_SEARCH_SERVICE_SECRET:-}"}"
export ALFRESCO_SECRET="${ALFRESCO_SECRET:-bitnami}"

[mdhont]

Thanks for the extra information. I will update the information in the task as well.

[rafariossaa]

Hi @victorcarabez ,
Now that you have the workaround, would you like to contribute and send a PR implementing those changes ? We will be glad to review and merge it.

[evaristorivi]

@victorcarabez

The second solution described in "On the other hand, for the communication between solr and alfresco to work correctly..." did not work for me if the first solution was not applied first, otherwise the error would indeed appear
Found 1 Solr nodes:
Solr process 1761 running on port 8983

so in my case I had to apply both solutions, not one of the two.

[gongomgra]

Hi @evaristorivi,

Thanks for sharing your solution. I'm sure it will be of help to other users in the future.

[github-actions]

This Issue has been automatically marked as "stale" because it has not had recent activity (for 15 days). It will be closed if no further activity occurs. Thanks for the feedback.

[github-actions]

Due to the lack of activity in the last 5 days since it was marked as "stale", we proceed to close this Issue. Do not hesitate to reopen it later if necessary.