From nikola@rs-pi.com Tue Dec 24 10:24:46 2002 Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 0BE78475843 for ; Tue, 24 Dec 2002 05:24:46 -0500 (EST) Received: from rs-pi.com (unknown [193.95.196.91]) by postgresql.org (Postfix) with ESMTP id 4C997474E61 for ; Tue, 24 Dec 2002 05:24:45 -0500 (EST) Received: from spot (fw.rs-pi.com [193.95.196.154]) by rs-pi.com (Postfix) with SMTP id 210258DE9 for ; Tue, 24 Dec 2002 11:29:42 +0100 (CET) Message-ID: <007d01c2ab36$85d92760$1100a8c0@rspi> From: "Nikola Ivacic" To: Subject: COPY command problems Date: Tue, 24 Dec 2002 11:23:38 +0100 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_007A_01C2AB3E.E79467A0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200212/266 X-Sequence-Number: 10854 This is a multi-part message in MIME format. ------=_NextPart_000_007A_01C2AB3E.E79467A0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: quoted-printable I need help Has somebody experienced difficulties using COPY command on large files. I have a large (250MB) file and each time I insert records I've got one or = more (< 30 of cca 1079000) corrupted records. The number of corrupted records is not constant (i.e. 1, 30, 7, 23= etc..) p.s. the file is tab-nl separated and corruption is done in the middle of r= ecord field (not the end or begining) lp Nikola ------=_NextPart_000_007A_01C2AB3E.E79467A0 Content-Type: text/html; charset="iso-8859-2" Content-Transfer-Encoding: quoted-printable
I need help
 
Has somebody experienced difficulties usin= g COPY=20 command on large files.
I have a large (250MB) file and each time = I insert=20 records I've got one or more (< 30 of cca 1079000)=20 corrupted
records. The number of corrupted records i= s not=20 constant (i.e. 1, 30, 7, 23 etc..)
 
p.s. the file is tab-nl separated and= =20 corruption is done in the middle of record field
(not the end or begining)
 
lp
 
Nikola
 
------=_NextPart_000_007A_01C2AB3E.E79467A0-- From tgl@sss.pgh.pa.us Tue Dec 24 15:24:46 2002 Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 7128A4762FE for ; Tue, 24 Dec 2002 10:24:46 -0500 (EST) Received: from sss.pgh.pa.us (unknown [192.204.191.242]) by postgresql.org (Postfix) with ESMTP id C468D4762D8 for ; Tue, 24 Dec 2002 10:24:45 -0500 (EST) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.12.6/8.12.6) with ESMTP id gBOFOnd0019756; Tue, 24 Dec 2002 10:24:50 -0500 (EST) To: "Nikola Ivacic" Cc: pgsql-sql@postgresql.org Subject: Re: COPY command problems In-reply-to: <007d01c2ab36$85d92760$1100a8c0@rspi> References: <007d01c2ab36$85d92760$1100a8c0@rspi> Comments: In-reply-to "Nikola Ivacic" message dated "Tue, 24 Dec 2002 11:23:38 +0100" Date: Tue, 24 Dec 2002 10:24:49 -0500 Message-ID: <19755.1040743489@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200212/267 X-Sequence-Number: 10855 "Nikola Ivacic" writes: > Has somebody experienced difficulties using COPY command on large files. > I have a large (250MB) file and each time I insert records I've got one or = > more (< 30 of cca 1079000) corrupted > records. The number of corrupted records is not constant (i.e. 1, 30, 7, 23= > etc..) I'd bet on flaky hardware --- have you run memory and disk tests? If the COPY data is passing across a network, then network problems are also worthy of suspicion. regards, tom lane From nikola@rs-pi.com Tue Dec 24 15:56:12 2002 Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id F03C5475A09 for ; Tue, 24 Dec 2002 10:56:12 -0500 (EST) Received: from rs-pi.com (unknown [193.95.196.91]) by postgresql.org (Postfix) with ESMTP id ADEF1475843 for ; Tue, 24 Dec 2002 10:56:11 -0500 (EST) Received: from spot (fw.rs-pi.com [193.95.196.154]) by rs-pi.com (Postfix) with SMTP id 60A5D8D8E; Tue, 24 Dec 2002 17:01:12 +0100 (CET) Message-ID: <013101c2ab64$d368df60$1100a8c0@rspi> From: "Nikola Ivacic" To: "Tom Lane" Cc: "Ales Kuclar" , References: <007d01c2ab36$85d92760$1100a8c0@rspi> <19755.1040743489@sss.pgh.pa.us> Subject: Re: COPY command problems Date: Tue, 24 Dec 2002 16:55:05 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200212/268 X-Sequence-Number: 10856 It must be internal error: two reasons: 1.) the original file is OK (I checked with grep + there is no network envolved) 2.) Error has strange patern: it substitutes 0x31 with 0x21 (1 with !) also 0x34 with 0x24 (4 with $) and 0x39 with 0x29 (9 with ) ) so I guess you are right. can you suggest some tools for FreeBSD to test RAM, because I think the hard disk is ok. p.s. right now I am testing it with splited file thanks Nikola ----- Original Message ----- From: "Tom Lane" To: "Nikola Ivacic" Cc: Sent: Tuesday, December 24, 2002 4:24 PM Subject: Re: [SQL] COPY command problems > "Nikola Ivacic" writes: > > Has somebody experienced difficulties using COPY command on large files. > > I have a large (250MB) file and each time I insert records I've got one or = > > more (< 30 of cca 1079000) corrupted > > records. The number of corrupted records is not constant (i.e. 1, 30, 7, 23= > > etc..) > > I'd bet on flaky hardware --- have you run memory and disk tests? If > the COPY data is passing across a network, then network problems are > also worthy of suspicion. > > regards, tom lane > From mallah@trade-india.com Fri Dec 27 11:12:53 2002 Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id BD508476029 for ; Fri, 27 Dec 2002 06:12:53 -0500 (EST) Received: from rtlocal.trade-india.com (mail-relay.trade-india.com [203.196.129.235]) by postgresql.org (Postfix) with SMTP id AAF5A475FC4 for ; Fri, 27 Dec 2002 06:12:49 -0500 (EST) Received: (qmail 1592 invoked from network); 27 Dec 2002 11:05:17 -0000 Received: from unknown (HELO system67.trade-india-local.com) (192.168.0.67) by infocom-236-129-del.trade-india.com with SMTP; 27 Dec 2002 11:05:17 -0000 From: "Rajesh Kumar Mallah." Organization: Infocom Network Limited. To: "Nikola Ivacic" , "Tom Lane" Subject: Re: COPY command problems Date: Fri, 27 Dec 2002 16:47:34 +0530 User-Agent: KMail/1.5 Cc: "Ales Kuclar" , References: <007d01c2ab36$85d92760$1100a8c0@rspi> <19755.1040743489@sss.pgh.pa.us> <013101c2ab64$d368df60$1100a8c0@rspi> In-Reply-To: <013101c2ab64$d368df60$1100a8c0@rspi> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200212271647.34446.mallah@trade-india.com> X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200212/272 X-Sequence-Number: 10860 http://www.memtest86.com/ may be useful. regds mallah. On Tuesday 24 December 2002 09:25 pm, Nikola Ivacic wrote: > It must be internal error: > two reasons: > 1.) the original file is OK (I checked with grep + there is no network > envolved) > 2.) Error has strange patern: it substitutes 0x31 with 0x21 (1 with !) also > 0x34 with 0x24 (4 with $) > and 0x39 with 0x29 (9 with ) ) > > so I guess you are right. > > can you suggest some tools for FreeBSD > to test RAM, because I think the hard disk is ok. > > p.s. right now I am testing it with splited file > > thanks > > Nikola > > ----- Original Message ----- > From: "Tom Lane" > To: "Nikola Ivacic" > Cc: > Sent: Tuesday, December 24, 2002 4:24 PM > Subject: Re: [SQL] COPY command problems > > > "Nikola Ivacic" writes: > > > Has somebody experienced difficulties using COPY command on large > > > files. I have a large (250MB) file and each time I insert records I've > > > got one > > or = > > > > more (< 30 of cca 1079000) corrupted > > > records. The number of corrupted records is not constant (i.e. 1, 30, > > > 7, > > 23= > > > > etc..) > > > > I'd bet on flaky hardware --- have you run memory and disk tests? If > > the COPY data is passing across a network, then network problems are > > also worthy of suspicion. > > > > regards, tom lane > > ---------------------------(end of broadcast)--------------------------- > TIP 2: you can get off all lists at once with the unregister command > (send "unregister YourEmailAddressHere" to majordomo@postgresql.org) -- Rajesh Kumar Mallah, Project Manager (Development) Infocom Network Limited, New Delhi phone: +91(11)6152172 (221) (L) ,9811255597 (M) Visit http://www.trade-india.com , India's Leading B2B eMarketplace.