After migrating the OS from RHEL-7.9 to RHEL-8 we are continuously receiving the below errors.
A system exception has occurred : Commit transaction failed: tperrno = 1, error text = <TPEABORT - transaction cannot commit>
A system exception has occurred : Enqueuing to QSpace [PPQ_SPACE1], Q [CPA] failed: tperrno = 13, [TPETIME - timeout occured]
A system exception has occurred : Enqueuing to QSpace [PPQ_SPACE1], Q [CPA] failed: tperrno = 12, [TPESYSTEM - internal system error]
Failed to open Resource Manager: tperrno = 16 error text = <TPERMERR - resource manager error
A system exception has occurred : Begin transaction failed: tperrno = 0 error text = <>
Tried to raise this issue with Tuxedo support and applied patch but still receiving this issues.
Related
I've developed an app in SwiftUI where I also use a chat with Parse LiveQuery.
The chat is working perfectly in real time until the app goes in background or the phone screen is closed.
If I receive a notification for a new message and I open the app from the notification then the LiveQuery stopes working and I recieve the error : "Parse LiveQuery: Error processing message: Optional(POSIXErrorCode: Software caused connection abort)"
I register the Message class
class Message: PFObject, PFSubclassing {
#NSManaged var message : String
#NSManaged var receiver : PFObject
#NSManaged var thread_id : PFObject
#NSManaged var produs : PFObject
#NSManaged var sender : PFObject
#NSManaged var picture : PFObject
class func parseClassName() -> String {
return "Message"
}
}
On the ChatView I have .onReceive where I start the function for chat and liveQuery
var messagesQuery: PFQuery<Message> {
return (Message.query()?
.whereKey("thread_id", equalTo: self.mainThread_obj)
.order(byDescending: "createdAt")) as! PFQuery<Message>
}
self.StartMessagesLiveQuery(messagesQuery: messagesQuery)
The func StartMessagesLiveQuery is:
let subscription: Subscription<Message> = Client.shared.subscribe(messagesQuery)
subscription.handle(Event.created) { query, object in
print("Live Query Trigered")
print(object)
DispatchQueue.global(qos: .background).async {
///reading object and appending to array}}
The only problem is when the app is opened from Notification.
I've there any way to force ParseLiveQuery to reconnect?
Error stack:
2021-03-03 10:15:47.442021+0200 App[35496:6350370] [connection] nw_read_request_report [C2] Receive failed with error "Software caused connection abort"
2021-03-03 10:15:47.449044+0200 App[35496:6350370] [connection] nw_read_request_report [C3] Receive failed with error "Software caused connection abort"
2021-03-03 10:15:47.449291+0200 App[35496:6350370] [connection] nw_read_request_report [C1] Receive failed with error "Software caused connection abort"
2021-03-03 10:15:47.455008+0200 App[35496:6350370] [connection] nw_flow_add_write_request [C2.1 52.1.38.170:443 failed channel-flow (satisfied (Path is satisfied), viable, interface: en0, ipv4, dns)] cannot accept write requests
2021-03-03 10:15:47.455055+0200 App[35496:6350370] [connection] nw_write_request_report [C2] Send failed with error "Socket is not connected"
2021-03-03 10:15:47.455815+0200 App[35496:6350657] ParseLiveQuery: Error processing message: Optional(POSIXErrorCode: Software caused connection abort)
2021-03-03 10:15:47.456612+0200 App[35496:6350370] Task <F6A87098-D5D3-444B-B656-D1999D5089D1>.<44> HTTP load failed, 42/0 bytes (error code: -1005 [1:53])
2021-03-03 10:15:47.462120+0200 App[35496:6350657] Task <F6A87098-D5D3-444B-B656-D1999D5089D1>.<44> finished with error [-1005] Error Domain=NSURLErrorDomain Code=-1005 "The network connection was lost." UserInfo={_kCFStreamErrorCodeKey=53, NSUnderlyingError=0x282d11830 {Error Domain=kCFErrorDomainCFNetwork Code=-1005 "(null)" UserInfo={NSErrorPeerAddressKey=<CFData 0x280174aa0 [0x20a814660]>{length = 16, capacity = 16, bytes = 0x100201bb344692570000000000000000}, _kCFStreamErrorCodeKey=53, _kCFStreamErrorDomainKey=1}}, _NSURLErrorFailingURLSessionTaskErrorKey=LocalDataTask <F6A87098-D5D3-444B-B656-D1999D5089D1>.<44>, _NSURLErrorRelatedURLSessionTaskErrorKey=(
"LocalDataTask <F6A87098-D5D3-444B-B656-D1999D5089D1>.<44>"
), NSLocalizedDescription=The network connection was lost., NSErrorFailingURLStringKey=https://site.b4a.app/classes/_User, NSErrorFailingURLKey=https://site.b4a.app/classes/_User, _kCFStreamErrorDomainKey=1}
2021-03-03 10:15:47.462237+0200 App[35496:6350657] [Error]: The network connection was lost. (Code: 100, Version: 1.19.1)
2021-03-03 10:15:47.462335+0200 Bazar[35496:6350657] [Error]: Network connection failed. Making attempt 1 after sleeping for 1.588128 seconds.
After this the network connection is restored and the app is functional, only the LiveQuery are not working.
I think i’ve solved it.
I've modified function StartMessagesLiveQuery:
func StartMessagesLiveQuery(messagesQuery: PFQuery<Message>){}
Inside the function I had:
let subscription: Subscription<Message> = Client.shared.subscribe(messagesQuery)
subscription!.handle(Event.created) { query, object in ... }
Now i’ve modified and put inside the Class
var subscription: Subscription<Message>?
var subscriber: ParseLiveQuery.Client!
and in the function i’ve modified with:
subscriber = ParseLiveQuery.Client()
subscription = subscriber.subscribe(messagesQuery)
I have added
ChromeOptions options = new ChromeOptions();
options.AddArgument("--ignore-certificate-errors-spki-list");
options.AddArgument("--ignore-ssl-errors");
options.AddArgument("test-type");
options.AddArguments("-incognito");
options.AddArgument("no-sandbox");
options.AddArgument("--start-maximized");
driver = new ChromeDriver(options);
But still getting:
ssl_client_socket_impl.cc(1061)] handshake failed error
How to suppress this error from console?
This error message...
[ERROR:ssl_client_socket_openssl.cc(855)] handshake failed; returned -1, SSL error code 1, net_error -2
...implies that the handshake failed between ChromeDriver and Chrome Browser failed at some point.
Root Cause
This error is generated due to net::SSLClientSocketImpl::DoHandshake and net::SSLClientSocketImpl implemented in ssl_client_socket_impl.cc
net::SSLClientSocketImpl::DoHandshake as follows:
int SSLClientSocketImpl::DoHandshake() {
crypto::OpenSSLErrStackTracer err_tracer(FROM_HERE);
int rv = SSL_do_handshake(ssl_.get());
int net_error = OK;
if (rv <= 0) {
int ssl_error = SSL_get_error(ssl_.get(), rv);
if (ssl_error == SSL_ERROR_WANT_CHANNEL_ID_LOOKUP) {
// The server supports channel ID. Stop to look one up before returning to
// the handshake.
next_handshake_state_ = STATE_CHANNEL_ID_LOOKUP;
return OK;
}
if (ssl_error == SSL_ERROR_WANT_X509_LOOKUP &&
!ssl_config_.send_client_cert) {
return ERR_SSL_CLIENT_AUTH_CERT_NEEDED;
}
if (ssl_error == SSL_ERROR_WANT_PRIVATE_KEY_OPERATION) {
DCHECK(ssl_config_.client_private_key);
DCHECK_NE(kSSLClientSocketNoPendingResult, signature_result_);
next_handshake_state_ = STATE_HANDSHAKE;
return ERR_IO_PENDING;
}
OpenSSLErrorInfo error_info;
net_error = MapLastOpenSSLError(ssl_error, err_tracer, &error_info);
if (net_error == ERR_IO_PENDING) {
// If not done, stay in this state
next_handshake_state_ = STATE_HANDSHAKE;
return ERR_IO_PENDING;
}
LOG(ERROR) << "handshake failed; returned " << rv << ", SSL error code "
<< ssl_error << ", net_error " << net_error;
net_log_.AddEvent(
NetLogEventType::SSL_HANDSHAKE_ERROR,
CreateNetLogOpenSSLErrorCallback(net_error, ssl_error, error_info));
}
next_handshake_state_ = STATE_HANDSHAKE_COMPLETE;
return net_error;
}
As per ERROR:ssl_client_socket_openssl.cc handshake failed the main issue is the failure of handshake when ChromeDriver handshakes with SSL pages in Chrome. Though Chromium team conducts test for SSL handshake through net_unittests, content_tests, and browser_tests but were not exhaustive. Some usecases are left out relying on the upstream tests.
Conclusion
This error won't interupt the execution of your Test Suite and you can ignore this issue for the time being till it is fixed by the Chromium Team.
You can restrict Chromium's log level to 3, so that only fatal errors are logged. Please bear in mind that you won't see any other error-related messages which might cause mayhem in production! The code looks like this:
var chromeOptions = new ChromeOptions();
chromeOptions.AddArgument("log-level=3");
var driver = new ChromeDriver(options : chromeOptions);
Chromium log levels are:
Description
Value
INFO
0
WARNING
1
LOG_ERROR
2
LOG_FATAL
3
When we set idle_in_transaction_session_timeout, the database will terminate connections that are idle for some time.
This works as expected, but I wonder how we should deal with this situation in the aplication code.
We are using pg-promise 10.3.1.
Details of the test:
we set the connection pool size to 1, so that we only have a single session
we set the for the idle-transaction-session-timeout to 2.5 sec:
SET idle_in_transaction_session_timeout TO 2500
now the active transaction will be in state idle in transaction:
see What can cause “idle in transaction” for “BEGIN” statements
now we start a transaction and sleep for 5 seconds
after 2.5sec the database will terminate the session and send an error to the client:
pgp-error error: terminating connection due to idle-in-transaction timeout
after another 2.5sec the transactional code tries to send a query (via the already terminated session), and this fails as expected:
dbIdle failed Error: Client has encountered a connection error and is not queryable
then pg-promise will try to rollback the transaction which will also fail (also expected, I guess)
But now we start a new query and also this query fails with
dbCall failed Client has encountered a connection error and is not queryable
is this expected? I was hoping that pg-promise can somehow remove the "broken" connection from the pool and that we could get a new one
obvously this is not the case, so how should we deal with this situation: i.e. how to recover, so that we can send new queries to the database?
Code example:
import pgPromise, { IMain } from "pg-promise";
import * as dbConfig from "./db-config.json";
import { IConnectionParameters } from "pg-promise/typescript/pg-subset";
const cll = "pg";
console.time(cll);
const pgp: IMain = pgPromise({
query(e) {
console.timeLog(cll,`> ${e.query}`);
},
error(e, ctx) {
console.timeLog(cll,"pgp-error", e);
}
});
const connectParams: IConnectionParameters = {
...dbConfig,
application_name: "pg-test",
max: 1
};
const db = pgp(connectParams);
/**
* #param timeoutMs 0 is no timeout
*/
async function setDbIdleInTxTimeout(timeoutMs: number = 0) {
await db.any("SET idle_in_transaction_session_timeout TO $1;", timeoutMs);
}
async function dbIdle(sleepTimeSec: number) {
console.timeLog(cll, `starting db idle ${sleepTimeSec}`);
const result = await db.tx(async t => {
await new Promise(resolve => setTimeout(resolve, sleepTimeSec * 1000));
return t.one("Select $1 as sleep_sec", sleepTimeSec);
});
console.timeLog(cll, result);
}
async function main() {
await setDbIdleInTxTimeout(2500);
try {
await dbIdle(5);
} catch (e) {
console.timeLog(cll, "dbIdle failed", e);
}
try {
await db.one("Select 1+1 as res");
} catch (e) {
console.timeLog(cll, "dbCall failed", e);
}
}
main().finally(() => {
pgp.end();
});
Console output (removed some useless lines):
"C:\Program Files\nodejs\node.exe" D:\dev_no_backup\pg-promise-tx\dist\index.js
pg: 23.959ms > SET idle_in_transaction_session_timeout TO 2500;
pg: 28.696ms starting db idle 5
pg: 29.705ms > begin
pg: 2531.247ms pgp-error error: terminating connection due to idle-in-transaction timeout
at TCP.onStreamRead (internal/stream_base_commons.js:182:23) {
name: 'error',
severity: 'FATAL',
code: '25P03',
}
pg: 2533.569ms pgp-error Error: Connection terminated unexpectedly
pg: 5031.091ms > Select 5 as sleep_sec
pg: 5031.323ms pgp-error Error: Client has encountered a connection error and is not queryable
pg: 5031.489ms > rollback
pg: 5031.570ms pgp-error Error: Client has encountered a connection error and is not queryable
pg: 5031.953ms dbIdle failed Error: Client has encountered a connection error and is not queryable
pg: 5032.094ms > Select 1+1 as res
pg: 5032.164ms pgp-error Error: Client has encountered a connection error and is not queryable
pg: 5032.303ms dbCall failed Error: Client has encountered a connection error and is not queryable
Process finished with exit code 0
This issue #680 has been fixed in pg-promise 10.3.5
How do I set the threadPool Size in slick ?
Here is my config:
pp {
profile = "slick.jdbc.OracleProfile$"
db {
url = "jdbc:oracle:thin:#52.4.90.244:1521:pp"
driver = oracle.jdbc.OracleDriver
keepAliveConnection = true
connectionPool = disabled
user = "xxx"
password = "xxx"
}
}
I am actually doing only one query, in streaming, do I really need a threadPool ? My goal is simply to retrieve all the rows from a table.
So far i am having the following error:
*** (s.basic.BasicBackend.stream) Signaling onSubscribe(slick.jdbc.JdbcBackend$JdbcStreamingActionContext#76f2bbc1)
*** (s.basic.BasicBackend.action) #1: [fused] cleanUp
try: andThen
1: PushStatementParameters StatementParameters(ForwardOnly,ReadOnly,null,null,100)
2: StreamingResultAction [select * FROM DRUG]
*** (s.basic.BasicBackend.action) #2: SynchronousDatabaseAction.Pin
*** (s.basic.BasicBackend.action) #3: [fused] andThen
1: PushStatementParameters StatementParameters(ForwardOnly,ReadOnly,null,null,100)
2: StreamingResultAction [select * FROM DRUG]
*** (s.basic.BasicBackend.action) #4: SynchronousDatabaseAction.Pin
*** (s.basic.BasicBackend.action) #5: SynchronousDatabaseAction.Unpin
*** (s.basic.BasicBackend.stream) Signaling onError(java.util.concurrent.RejectedExecutionException: Task slick.basic.BasicBackend$DatabaseDef$$anon$2#3061964c rejected from slick.util.AsyncExecutor$$anon$2$$anon$1#55d260cc[Shutting down, pool size = 1, active threads = 1, queued tasks = 0, completed tasks = 0])
An internal error occurred during: "Building workspace".
jzentry == 0,
jzfile = 127476560,
total = 28,
name = C:\Users\amar.agarwal5\.m2\repository\org\apache\maven\plugins\maven-antrun-plugin\1.6\maven-antrun-plugin-1.6.jar,
i = 1,
message = null
what should i do??
I have just downloaded the hibernate tools from marketplace and after that I am getting this error on restart