Date: Sat, 6 Oct 2001 21:52:27 +0200 (CEST) From: "Hartmann, O." <ohartman@klima.physik.uni-mainz.de> To: <freebsd-ports@freebsd.org> Subject: MySQL server problems since FBSD 4.4! Message-ID: <20011006213612.X4043-100000@klima.physik.uni-mainz.de>
next in thread | raw e-mail | index | archive | help
Dear Sirs. Since the switch from FBSD 4.3 towards 4.4, MySQL servers 3.23.XX has a problem I can not fix. I need help or someone to discuss how to target the problem! Background: we use a simple table for our dayly climate measurements. Field are: id (timestamp), zeit (English 'time', type datetime, primary key) and 16 fields of double to keep data. Insertion is done via a small self written C programm, that scans, disect and inserts data into the table by checking first, whether date/time is present or not. Via a simple debug routine I can let the program show which fields contain data and are about to be inserted, and from the view of the C program everything seems clear. Insertion is done via a simple call of the API function from MySQL library. Phenomenon: After a reboot of the machine insertion seems to fail on all data columns (type 'DOUBLE'), only the id (TIMESTAMP) and the date/time column has valid data, all data columns have 'NaN' althought the C-program disect all data the correct way and store them into variables of type 'float' (by the way: everytime we upgraded our MySQL server our software has been recompiled to hit the newest libraries!). The only wy to get rid of these faulty 'NaN' entries is to delete them from the table, then call the C-program again, look whether the faulty values have been inserted again, if yes, delete them ... until the right values get inserted. From this point on the system works as expected. This phenomenon does not occur with FreeBSD 4.3-STABLE and all tested MySQL servers (our actual server is MySQL 3.23.42 from the ports collection). This phenomenon seems to me like a bug either in MySQL or FreeBSD, but the fact that it occurs in all MySQL servers with FreeBSD 4.4 but not in FreeBSD 4.3 makes me sure it is a kind of FreeBSD bug. The same time, when 'moving' from 4.3 to 4.4 and in parallel to the first occurance of this phenomenon I have had on all FreeBSD 4.4-RC and RELEASE machines later on the phenomenon, that looking at network traffic with netstat and reporting bandwith with a FTP client, shows 'NaN' instead of the expected 'KB/s-value' or transmitted bytes. And this made me suspecting FreeBSD. Is someone out here watching similar problems? So far, the time, the SQL server is running its all right, the system is stable and not about to be 'faulty', but after a machine's reboot (not a simple MySQL server shutdown and restart, this has no negative effect). Thanks in advance, Oliver -- MfG O. Hartmann ohartman@klima.physik.uni-mainz.de ---------------------------------------------------------------- IT-Administration des Institutes fuer Physik der Atmosphaere (IPA) ---------------------------------------------------------------- Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinenraum) Tel: +496131/3924144 FAX: +496131/3923532 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011006213612.X4043-100000>