Hey Jasper , we have done several levels of integration testing on deposits / account management services. If you would like to have our notes shared to you let me know.
Conclusion of our study 1. Deadlocks are appearing since the commands are not able to fulfill some requests. We realised there were some Code types missing , like in our case we were trying to make a deposit in an account. 2. Rhythm service does not work as expected due to the ephemeral nature of the command service in between. 3. Sometimes Cassandra does not process these transactions because the previous ones have failed or the code written in the services as of now the code base does not do anything. We were left wondering what was the issue for several weeks. let me know if you have any questions On Fri, Sep 4, 2020 at 9:31 PM Ed Cable <[email protected]> wrote: > Hi Jasper, > > Welcome to the community. I'm glad to see another organization > beginning to adopt Fineract CN. The wallet use case is the primary one > we're trying to advanced as a community initially around Fineract CN so > thank you for your transparency and willingness to collaborate on list. I'm > looking forward to learning more of what Apollo is building and how we can > assist you in contributing back to the upstream, and joining the concurrent > working groups we have going around wallet requirements and performance and > scalability. > > I'll try to flag this thread to some of the others in the performance and > scalability working group to give their inputs. > > @Victor Manuel Romero Rodriguez <[email protected]> @Jose > Hernandez <[email protected]> Based on the performance > testing you have done on Fineract-CN to date, can you give some feedback to > Jasper and his team? > > Thanks, > > Ed > > On Fri, Sep 4, 2020 at 12:40 AM <[email protected]> wrote: > >> Good day to all. >> >> Currently I am working on simulating concurrent debit and credit >> transactions (similar to a wallet environment) via journal entries. My >> setup are as follows >> >> Load Test: >> - JMeter with third party plugin named Ultimate Thread Group to simulate >> parallel requests across users >> - Ultimate Thread Group parameters >> - Start Threads Count: 10 >> - Initial Delay: 10s >> - Startup Time: 30s >> - Hold Time: 30s >> - Shutdown Time: 30s >> - Ultimate Thread Group has 3 samplers which all are HTTP Requests >> - 1 sampler is a login request with a once only controller to fetch the >> accessToken to be passed on the next two requests >> - 2 samplers which are debit and credit requests via addJournal API call >> with a transfer amount of 1.00 >> >> Fineract CN: >> - Setup of Fineract CN is based on >> https://github.com/apache/fineract-cn-docker-compose >> - Created a Liability-type ledger with identifier 5000 >> - Created a subledger with identifier 5200 named as First Wallet Group >> - Created two wallets (accounts) under 5200 with names Wallet 1, Wallet >> 2, and identifiers 5201, 5202, respectively having both 50,000 balance >> each. >> >> Platform and Specifications: >> - Google Cloud Platform - 8vCPUs, 30GB memory, 100GB SSD >> >> After the load test, statistics shown that there are 92 credit >> transactions and 84 debit transactions for the account 5201. The >> balances of both wallets are found below: >> Wallet 1 (5201): 50,016 >> Wallet 2 (5202): 49,999 >> >> My concerns are first, there are deadlocks occurring upon load testing. >> Second, it appears there is a balance discrepancy where some of the >> credit and debit transactions >> are missing for both accounts. Computed balance from the above >> statistics shows the balance of 5201 should be 50,008 and 5202 being >> 49,992. Even after load testing, balance >> of both accounts should remain the same as it were initially upon >> account creation, but that is not the case. Third, I have checked in the >> cassandra db where there are 113 >> pending and 63 processed transactions, but it seems the 113 does not >> match with the results above where most transactions are successfully >> transferred/processed. What I would like to know is that if I have >> missed something upon setting up or any configurations that must be >> tweaked before proceeding to such testing. I greatly appreciate to the >> help I will be given. >> >> Best regards, >> Jasper >> > > > -- > *Ed Cable* > President/CEO, Mifos Initiative > [email protected] | Skype: edcable | Mobile: +1.484.477.8649 > > *Collectively Creating a World of 3 Billion Maries | *http://mifos.org > <http://facebook.com/mifos> <http://www.twitter.com/mifos> > > -- Thanks and regards, Saransh Sharma Research Partner This mail is governed by Muellners® IT policy. The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message, including any attachments. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. All messages sent to and from this e-mail address may be monitored as permitted by applicable law and regulations to ensure compliance with our internal policies and to protect our business. E-mails are not secure and cannot be guaranteed to be error free as they can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by e-mail.
