From the Oracle Database 21 documentation:https://docs.oracle.com/en/database/oracle/oracle-database/21/refrn/JOB_QUEUE_PROCESSES.html “In a CDB root: Set JOB_QUEUE_PROCESSES to the maximum number of job slaves that can be used simultaneously in the entire CDB. Oracle recommends that you set the value of this parameter to at least twice the number of open containers in the CDB, otherwise, there might be severe starvation between PDBs trying to run multiple jobs. If JOB_QUEUE_PROCESSES is set to 0 in a CDB root, then DBMS_JOB and Oracle Scheduler jobs cannot run in the CDB root or in any PDB, regardless of the JOB_QUEUE_PROCESSES setting at the PDB level. “In a PDB: Set JOB_QUEUE_PROCESSES to the maximum number of job slaves that can be used simultaneously in the PDB. The actual number depends on the resources assigned by Resource Manager and the demand in other containers. When multiple PDBs request jobs, Oracle Scheduler attempts to give all PDBs a fair share of the processes. Oracle
MOS 2118028.1 tries to explain this, but they only cover the zero case: If it's zero in the CDB, then the scheduler is disabled in all PDBs, but if it's zero in the PDB, it's only disabled in that one particular PDB. It looks like the CDB value works as a ceiling. If that’s true, I think we should set the CDB value astronomically high, like the Oracle default value of 4000, then fine tune the PDB value. If anyone disagrees, I would like an in-depth explanation for why. I may come back here with more details if I get a chance to test this more fully.
Apps9 - UPD deploying failed on the customer non-prod environment. Work without error on customer Apps9 internal environment. If you use DataPump as the SYSTEM user, it won’t include grants on SYS objects and won’t include system privileges. I recommend doing nonproduction copies with RMAN instead.
I’ve found that background jobs can hide a lot of errors whenever TRANSACTION_SYS shows up in the error stack, so my first thought here is usually to run it live. However, this particular procedure detects whether it’s being run as a background job and will spawn itself as a job if it isn’t. This makes troubleshooting annoying because you can’t run it live from the GUI. If you have access to the app owner, you might try this hacky workaround in Oracle. The idea is to make it think it’s running as a background job so it runs live, selecting a negative background job ID that will never exist.(Using a raw DELETE query like this is not considered a best practice, so if you can, it would be best if you could refresh a nonproduction environment and try it there.)BEGIN DELETE FROM transaction_sys_local_tab WHERE id = -1; DELETE FROM transaction_sys_status_tab WHERE id = -1; transaction_sys.set_current_job_id(-1); mpccom_accounting_api.transfer_to_finance ( contract_ =>
Try Sigsync Office 365 email Signature, Its server side email signature Server-side email signatures don’t give end users the option to change theirs. You might think this is a good thing, like if someone puts something objectionable, controversial, or inappropriate in their signature, but I’ve been around IT long enough to recognize there will be several problems with not giving this power to end users.It’s going to annoy employees, and annoying your highest performers is a great way to push them out to a different company that trusts them to make this kind of decision for themselves. This doesn’t give you the power you think it does. The message body might ALSO contain unwanted content, or end users might pad the bottom of their message with the start of their custom signature. Signatures will get duplicated. End users will think it’s weird to send a message with no signature, so they’ll add one, and then your customers will get emails with two signatures. I see this from
I’m very disappointed in this regression. DNS is a nonnegotiable requirement for managing an enterprise-grade network, and every exception that doesn’t use a FQDN adds complexity to the total configuration, requiring extra time or tribal knowledge to troubleshoot or change later.What is the impact of setting ifscore.networkpolicy.enabled to false, in more specific technical language?Where in the code stack is this happening? What might a good starting point be for developing a better workaround?
The maximum length of an employee is hard-coded all over Ifs.Application.Person.dll and probably many others. I doubt we’ll want to take that risk.
There’s no technical reason to limit a VARCHAR2 like this. The length is really just a constraint. It only uses the room it needs for the data. -- DROP TABLE c_short_tab PURGE;-- DROP TABLE c_long_tab PURGE;CREATE TABLE c_short_tab (x VARCHAR2(11)) TABLESPACE ifsapp_data;CREATE TABLE c_long_tab (x VARCHAR2(100)) TABLESPACE ifsapp_data;DECLARE hash_ RAW(32); value_ VARCHAR2(11);BEGIN hash_ := HEXTORAW('0'); FOR i IN 0..1048575 LOOP hash_ := dbms_crypto.hash(hash_, dbms_crypto.hash_sh256); value_ := SUBSTR(RAWTOHEX(hash_), 1, 11); INSERT INTO c_short_tab (x) VALUES (value_); INSERT INTO c_long_tab (x) VALUES (value_); END LOOP;END;/COMMIT;SELECT segment_name, bytes, blocks, extents FROM user_segments WHERE segment_name IN ('C_LONG_TAB', 'C_SHORT_TAB'); SEGMENT_NAME BYTES BLOCKS EXTENTS C_LONG_TAB 20971520 2560 35 C_SHORT_TAB 20971520 2560 35
What is the purpose of making it 12 characters? For years, our standard practice was to create a user ID with 8 characters, but this broke when two people had the same name. We switched to a 12-character system to resolve name conflicts, and that worked during testing. We don’t use the HR module, and only users that need to book time against projects are created as employees, so now this edge case has left us stuck. If a user ID can be 30 characters and a person ID can be 20, it makes no sense to me why an employee ID is limited to 11. What was the point of limiting that so tightly?
The .net client field has a max length so you need to change that. Java code might be affected depending on which field it is. If it is used in integrations or other things where we use Java like in DOCMAN. It’s the employee ID. It’s so thoroughly littered everywhere, I figured it might be easier and safer to change the length of every 11-character column to 12 characters, doing something like this… (I haven’t tested any of this yet. It’s late on a Sunday night here.) BEGIN FOR rec_ IN ( SELECT 'ALTER TABLE "' || table_name || '" ' || 'MODIFY "' || column_name || '" ' || 'VARCHAR2(12)' AS sq FROM user_tab_cols WHERE table_name IN (SELECT table_name FROM user_tables WHERE tablespace_name IS NOT NULL) AND table_name NOT IN (SELECT mview_name FROM user_mviews) AND table_name NOT IN (SELECT queue_table FROM user_queues) AND table_name NOT LIKE 'AQ$%' AND data_ty
Cross posted as support case CS0147101
Has this answer changed in 23R1?
You can do this kind of migration work by reverse-engineering the web service calls, whether from your web browser’s tools or from the Aurena developer console. I wrote a working solution to your problem above here in PowerShell. It uses basic authentication against the IFSADMIN user. You can remove the Get-Projection function if you’d like.$basic_username = 'IFSADMIN'$basic_password = 'ifsadmin_password'$hostname = 'ifs.example.com'$credential_plaintext = "$($basic_username):$($basic_password)"$credential_bytes = [System.Text.Encoding]::UTF8.GetBytes($credential_plaintext)$credential_base64 = [Convert]::ToBase64String($credential_bytes)$headers = New-Object "System.Collections.Generic.Dictionary[[String],[String]]"$headers.Add("Content-Type", "application/json")$headers.Add("Authorization", "Basic $($credential_base64)")Function Post-Projection { param($service_and_set, $body_hashtable) $uri = 'https://' + $global:hostname + '/main/ifsapplications/projection/v1/' + $servi
Our system has over 100K columns and 6K tables, and the Oracle EE pack would require a prohibitive amount of data entry into Enterprise Manager.
I’m also considering hash partitioning the remaining indexes. This article has some really good diagrams to explain why:https://uhesse.com/2016/12/02/how-to-reduce-buffer-busy-waits-with-hash-partitioned-indexes-in-oracle/Normally as you’re writing new history log entries, you’ll fill in values on the very right side of the B+tree, and concurrent transactions are going to fight for the same leaf blocks as they write the history entries with almost identical values…1000001100000210000031000004... (Copying the image from that article)Hashing the index key lets us pseudo-randomize a bucket in which to partition the index, reducing contention for the hot blocks: Another common solution is to reverse the key, but that would make the indexes useless for sorts and range scans. At least with partitioning, the optimizer stands a good chance reading the necessary records off of each partition.
Even with the schema remapping and the data repairs, you’ll also run into CONTEXT objects that will need to be recreated. DROP CONTEXT "ABSCALCALL_CTX";DROP CONTEXT "ABSENCEACTIONSTATUS_CTX";DROP CONTEXT "ABSENCEACTIONTYPE_CTX";DROP CONTEXT "ABSENCEBASE_CTX";-- (These are just 4 examples of the 2,679 of these in our environment.)CREATE OR REPLACE CONTEXT "ABSCALCALL_CTX" USING "IFSAPP"."DOMAIN_SYS" ACCESSED GLOBALLY;CREATE OR REPLACE CONTEXT "ABSENCEACTIONSTATUS_CTX" USING "IFSAPP"."DOMAIN_SYS" ACCESSED GLOBALLY;CREATE OR REPLACE CONTEXT "ABSENCEACTIONTYPE_CTX" USING "IFSAPP"."DOMAIN_SYS" ACCESSED GLOBALLY;CREATE OR REPLACE CONTEXT "ABSENCEBASE_CTX" USING "IFSAPP"."DOMAIN_SYS" ACCESSED GLOBALLY;
Your DataPump import file should remap the schema.REMAP_SCHEMA=ACME1APP:IFSAPP
I got the binary ZIP upload working. # I tested this in PowerShell 7.3.# JSON behavior might be a little different if you're on a different version.Install-Module powershell-yaml$ifscloud_values_path = '\\ims_server.example.com\c$\IFS\PRD\UB1-MWS-2213UB\ifsroot\config\ifscloud-values.yaml'$ifscloud_values = Get-Content -Path $ifscloud_values_path -Raw$ifscloud_values = $ifscloud_values | ConvertFrom-Yaml$basic_username = 'IFSADMIN'$basic_password = $ifscloud_values.ifscore.passwords.ifsadminpassword.data$credential_plaintext = "$($basic_username):$($basic_password)"$credential_bytes = [System.Text.Encoding]::UTF8.GetBytes($credential_plaintext)$credential_base64 = [Convert]::ToBase64String($credential_bytes)$hostname = $ifscloud_values.global.systemurl$headers = New-Object "System.Collections.Generic.Dictionary[[String],[String]]"$headers.Add("Content-Type", "application/json")$headers.Add("Authorization", "Basic $($credential_base64)")$global:hostname = $hostname$global:headers = $hea
If I may make a public recommendation, it would be to encourage every software employee, working at IFS or elsewhere, to expand outside their bubble just a little bit and continually cross-train in something new every week or month. Sure, specialization provides incredible efficiencies, where everyone gets into their groove doing what they do best, but the handoffs from one specialist to another eat into those efficiencies. If a popular department gets overloaded, it slows down everyone if employees can’t adapt, slowing down the customer experience. I’m not asking for anyone to become a jack of all trades, master of none, but branching out a little helps not only your customer experience today but also makes your résumé far more marketable tomorrow.
I’ve been developing software as a hobby since the 90s and professionally since the early 2000s. I’ve been using IFS since 2008 when I was a mechanical engineer and have been doing serious IFS development since about 2011. I’ve had a lot of trouble with IFS providing deliveries on time that function correctly the first time, and every one of these problems has stemmed from their development and delivery practices and the disconnect between IFS Global and IFS North America. North America used to recommend a nonstandard application owner like ACME1APP. The credit card, freight, and tax interfaces, the Italian finance pack in Apps 8, and the Global Extension in Apps 9 were all sold as bolt-on, off-the-shelf components, but each of these has required support as though they were customizations, with sticky tentacles reaching all throughout the core code, breaking the entire application if any one of these components wasn’t deployed perfectly. For the first time in a very long time, I do see
Someone PM’ed me with a followup question about this. The error they were getting was ORA-29283. The question had a a file share with a format that looks like \\SERVER\SHARE. (See footnote 📂.) I'm guessing that your files are on a Windows server. Consequently, your Oracle database will also need to be running on Windows for this to be doable without a lot of hacking. (See footnote🐧.) I highly recommend Don Burleson’s website for all things Oracle.http://www.dba-oracle.com/t_ora_29283_invalid_file_operation.htmIn short, he says this is an OS issue. Unfortunately, because so much of his experience has been with UNIX and UNIX-like operating systems, you won’t find as much help there for Windows-specific issues. (There is some, just not as much.) The process oracle.exe for your Oracle Database service is going to be running as some user. If you haven't messed around with it during installation, it's probably running as the SYSTEM user. This is the user that will need access to the file
I wish IFS Community allowed us to offer a bounty on a hard, critical question, like they do on a certain other Q&A website starting with the letters S.O. 😉
Any thoughts on this?
Are you sure the wildcard is not honored if the status is Generating? That's a big that should be reported, if it is the case. Yes.Thank you.
Bump
Already have an account? Login
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.
