Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
Hi,all I’m considering removing the RequiresInstanceClassLoading annotation from class AbstractHadoopProcessor, Does anybody know the potential Influence? Thanks By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
Hi Hai Luo, @RequiresInstanceClassLoading makes possible to configure separate / isolated "Additional Classpath Resources" settings on your HDFS processors (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). Is there any specific reason / use case why you are considering to remove it? Regards, Peter Turcsanyi On Mon, Nov 11, 2019 at 3:30 PM abellnotring wrote: > Hi,all > I’m considering removing the RequiresInstanceClassLoading annotation > from class AbstractHadoopProcessor, > Does anybody know the potential Influence? > > Thanks > By Hai Luo >
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
Peter The most common challenge is if two isolated instances both want to use a native lib. No two native libs with the same name can be in the same jvm. We need to solve that for sure. Thanks On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi wrote: > Hi Hai Luo, > > @RequiresInstanceClassLoading makes possible to configure separate / > isolated "Additional Classpath Resources" settings on your HDFS processors > (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). > > Is there any specific reason / use case why you are considering to remove > it? > > Regards, > Peter Turcsanyi > > On Mon, Nov 11, 2019 at 3:30 PM abellnotring > wrote: > >> Hi,all >> I’m considering removing the RequiresInstanceClassLoading annotation >> from class AbstractHadoopProcessor, >> Does anybody know the potential Influence? >> >> Thanks >> By Hai Luo >> >
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
I’m assuming your talking about the snappy problem. If you use compress content prior to puthdfs you can compress with Snappy as it uses the Java Native Snappy Lib. The HDFS processors are limited to the actual Hadoop Libraries so they’d have to change from Native to get around this. I’m pretty sure we need instance loading to handle the other issues mentioned. Thanks Shawn From: Joe Witt Reply-To: "users@nifi.apache.org" Date: Monday, November 11, 2019 at 8:56 AM To: "users@nifi.apache.org" Subject: Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor Peter The most common challenge is if two isolated instances both want to use a native lib. No two native libs with the same name can be in the same jvm. We need to solve that for sure. Thanks On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi mailto:turcsa...@apache.org>> wrote: Hi Hai Luo, @RequiresInstanceClassLoading makes possible to configure separate / isolated "Additional Classpath Resources" settings on your HDFS processors (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). Is there any specific reason / use case why you are considering to remove it? Regards, Peter Turcsanyi On Mon, Nov 11, 2019 at 3:30 PM abellnotring mailto:abellnotr...@sina.com>> wrote: Hi,all I’m considering removing the RequiresInstanceClassLoading annotation from class AbstractHadoopProcessor, Does anybody know the potential Influence? Thanks By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
I can’t remember for sure but I think if you use CompressContent the compressed file has to fit in a single HDFS file block in order to work. IIRC Hadoop-Snappy is different from regular Snappy in the sense that it puts the compression header in each block so the file can be reassembled and decompressed correctly. > On Nov 11, 2019, at 10:30 AM, Shawn Weeks wrote: > > > I’m assuming your talking about the snappy problem. If you use compress > content prior to puthdfs you can compress with Snappy as it uses the Java > Native Snappy Lib. The HDFS processors are limited to the actual Hadoop > Libraries so they’d have to change from Native to get around this. I’m pretty > sure we need instance loading to handle the other issues mentioned. > > Thanks > Shawn > > From: Joe Witt > Reply-To: "users@nifi.apache.org" > Date: Monday, November 11, 2019 at 8:56 AM > To: "users@nifi.apache.org" > Subject: Re: Influence about removing RequiresInstanceClassLoading from > AbstractHadoopProcessor processor > > Peter > > The most common challenge is if two isolated instances both want to use a > native lib. No two native libs with the same name can be in the same jvm. > We need to solve that for sure. > > Thanks > > On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi wrote: > Hi Hai Luo, > > @RequiresInstanceClassLoading makes possible to configure separate / isolated > "Additional Classpath Resources" settings on your HDFS processors (eg. S3 > storage driver on one of your PutHDFS and Azure Blob on the other). > > Is there any specific reason / use case why you are considering to remove it? > > Regards, > Peter Turcsanyi > > On Mon, Nov 11, 2019 at 3:30 PM abellnotring wrote: > Hi,all > I’m considering removing the RequiresInstanceClassLoading annotation > from class AbstractHadoopProcessor, > Does anybody know the potential Influence? > > Thanks > By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
Hi,Peter & All I’m using kylo to manage the nifi flow(called feed in kylo),and there are 4200 instances(600+ instances extended from AbstractHadoopProcessor) in my nifi canvas,The NIFI Non-Heap Memory has increased more than 6GB after some days running ,which is extremely abnormal . I have analyzed the class loaded into Compressed Class Space ,and found most of the CCS was used by classes related by AbstractHadoopProcessor. So I think removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor may be a Solution for reducing the CCS used. Do you have any ideas for this ? Thanks By Hai Luo On 11/12/2019 02:17,Shawn Weeks wrote: I’m assuming your talking about the snappy problem. If you use compress content prior to puthdfs you can compress with Snappy as it uses the Java Native Snappy Lib. The HDFS processors are limited to the actual Hadoop Libraries so they’d have to change from Native to get around this. I’m pretty sure we need instance loading to handle the other issues mentioned. Thanks Shawn From: Joe Witt Reply-To: "users@nifi.apache.org" Date: Monday, November 11, 2019 at 8:56 AM To: "users@nifi.apache.org" Subject: Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor Peter The most common challenge is if two isolated instances both want to use a native lib. No two native libs with the same name can be in the same jvm. We need to solve that for sure. Thanks On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi <turcsa...@apache.org> wrote: Hi Hai Luo, @RequiresInstanceClassLoading makes possible to configure separate / isolated "Additional Classpath Resources" settings on your HDFS processors (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). Is there any specific reason / use case why you are considering to remove it? Regards, Peter Turcsanyi On Mon, Nov 11, 2019 at 3:30 PM abellnotring <abellnotr...@sina.com> wrote: Hi,all I’m considering removing the RequiresInstanceClassLoading annotation from class AbstractHadoopProcessor, Does anybody know the potential Influence? Thanks By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
If you remove the @RequiresInstanceClassloading, the UserGroupInformation class from Hadoop (hadoop-common, if I remember correctly) will be shared across all instances that come from a particular NAR (such as PutHDFS, ListHDFS, FetchHDFS, etc, from nifi-hadoop-nar-x.y.z.nar). If you are using Kerberos in those processors and configured different principals across the various processors, you could run into issues when the processors attempt to acquire new TGTs, most likely the first time a relogin is attempted. UGI has some static state and @RequiresInstanceClassloading makes sure each instance of a processor with that annotation has its own classloader to keep that kind of state from being shared across instances. On Mon, Nov 11, 2019 at 9:41 PM abellnotring wrote: > Hi,Peter & All > I’m using kylo to manage the nifi flow(called feed in kylo),and there > are 4200 instances(600+ instances extended from AbstractHadoopProcessor) > in my nifi canvas,The NIFI Non-Heap Memory has increased more than 6GB > after some days running ,which is extremely abnormal . I have analyzed the > class loaded into Compressed Class Space ,and found most of the CCS was > used by classes related by AbstractHadoopProcessor. >So I think removing RequiresInstanceClassLoading from > AbstractHadoopProcessor processor may be a Solution for reducing the CCS > used. >Do you have any ideas for this ? > > > <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> > > > <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> > > > <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> > <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> > Thanks > > > By Hai Luo > On 11/12/2019 02:17,Shawn Weeks > wrote: > > I’m assuming your talking about the snappy problem. If you use compress > content prior to puthdfs you can compress with Snappy as it uses the Java > Native Snappy Lib. The HDFS processors are limited to the actual Hadoop > Libraries so they’d have to change from Native to get around this. I’m > pretty sure we need instance loading to handle the other issues mentioned. > > > > Thanks > > Shawn > > > > *From: *Joe Witt > *Reply-To: *"users@nifi.apache.org" > *Date: *Monday, November 11, 2019 at 8:56 AM > *To: *"users@nifi.apache.org" > *Subject: *Re: Influence about removing RequiresInstanceClassLoading from > AbstractHadoopProcessor processor > > > > Peter > > > > The most common challenge is if two isolated instances both want to use a > native lib. No two native libs with the same name can be in the same jvm. > We need to solve that for sure. > > > > Thanks > > > > On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi > wrote: > > Hi Hai Luo, > > > > @RequiresInstanceClassLoading makes possible to configure separate / > isolated "Additional Classpath Resources" settings on your HDFS processors > (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). > > > > Is there any specific reason / use case why you are considering to remove > it? > > > > Regards, > > Peter Turcsanyi > > > > On Mon, Nov 11, 2019 at 3:30 PM abellnotring > wrote: > > Hi,all > > I’m considering removing the RequiresInstanceClassLoading annotation > from class AbstractHadoopProcessor, > > Does anybody know the potential Influence? > > > > Thanks > > By Hai Luo > >
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
What version of NiFi is Kylo based on? There was a memory leak related to HDFS processors that was fixed back in 1.7.0: https://issues.apache.org/jira/browse/NIFI-5136 On Mon, Nov 11, 2019 at 11:14 PM Jeff wrote: > > If you remove the @RequiresInstanceClassloading, the UserGroupInformation > class from Hadoop (hadoop-common, if I remember correctly) will be shared > across all instances that come from a particular NAR (such as PutHDFS, > ListHDFS, FetchHDFS, etc, from nifi-hadoop-nar-x.y.z.nar). If you are using > Kerberos in those processors and configured different principals across the > various processors, you could run into issues when the processors attempt to > acquire new TGTs, most likely the first time a relogin is attempted. UGI has > some static state and @RequiresInstanceClassloading makes sure each instance > of a processor with that annotation has its own classloader to keep that kind > of state from being shared across instances. > > On Mon, Nov 11, 2019 at 9:41 PM abellnotring wrote: >> >> Hi,Peter & All >> I’m using kylo to manage the nifi flow(called feed in kylo),and there >> are 4200 instances(600+ instances extended from AbstractHadoopProcessor) in >> my nifi canvas,The NIFI Non-Heap Memory has increased more than 6GB after >> some days running ,which is extremely abnormal . I have analyzed the class >> loaded into Compressed Class Space ,and found most of the CCS was used by >> classes related by AbstractHadoopProcessor. >>So I think removing RequiresInstanceClassLoading from >> AbstractHadoopProcessor processor may be a Solution for reducing the CCS >> used. >>Do you have any ideas for this ? >> >> >> >> Thanks >> >> >> By Hai Luo >> >> On 11/12/2019 02:17,Shawn Weeks wrote: >> >> I’m assuming your talking about the snappy problem. If you use compress >> content prior to puthdfs you can compress with Snappy as it uses the Java >> Native Snappy Lib. The HDFS processors are limited to the actual Hadoop >> Libraries so they’d have to change from Native to get around this. I’m >> pretty sure we need instance loading to handle the other issues mentioned. >> >> >> >> Thanks >> >> Shawn >> >> >> >> From: Joe Witt >> Reply-To: "users@nifi.apache.org" >> Date: Monday, November 11, 2019 at 8:56 AM >> To: "users@nifi.apache.org" >> Subject: Re: Influence about removing RequiresInstanceClassLoading from >> AbstractHadoopProcessor processor >> >> >> >> Peter >> >> >> >> The most common challenge is if two isolated instances both want to use a >> native lib. No two native libs with the same name can be in the same jvm. >> We need to solve that for sure. >> >> >> >> Thanks >> >> >> >> On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi wrote: >> >> Hi Hai Luo, >> >> >> >> @RequiresInstanceClassLoading makes possible to configure separate / >> isolated "Additional Classpath Resources" settings on your HDFS processors >> (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). >> >> >> >> Is there any specific reason / use case why you are considering to remove it? >> >> >> >> Regards, >> >> Peter Turcsanyi >> >> >> >> On Mon, Nov 11, 2019 at 3:30 PM abellnotring wrote: >> >> Hi,all >> >> I’m considering removing the RequiresInstanceClassLoading annotation >> from class AbstractHadoopProcessor, >> >> Does anybody know the potential Influence? >> >> >> >> Thanks >> >> By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
I’ve uploaded lots files to HDFS that were bigger and smaller than a block. Last time I tried compression with PutHDFS I didn’t notice any block splitting happening either. It still showed up a single file. Thanks From: Matt Burgess Reply-To: "users@nifi.apache.org" Date: Monday, November 11, 2019 at 10:13 AM To: "users@nifi.apache.org" Subject: Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor I can’t remember for sure but I think if you use CompressContent the compressed file has to fit in a single HDFS file block in order to work. IIRC Hadoop-Snappy is different from regular Snappy in the sense that it puts the compression header in each block so the file can be reassembled and decompressed correctly. On Nov 11, 2019, at 10:30 AM, Shawn Weeks wrote: I’m assuming your talking about the snappy problem. If you use compress content prior to puthdfs you can compress with Snappy as it uses the Java Native Snappy Lib. The HDFS processors are limited to the actual Hadoop Libraries so they’d have to change from Native to get around this. I’m pretty sure we need instance loading to handle the other issues mentioned. Thanks Shawn From: Joe Witt Reply-To: "users@nifi.apache.org" Date: Monday, November 11, 2019 at 8:56 AM To: "users@nifi.apache.org" Subject: Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor Peter The most common challenge is if two isolated instances both want to use a native lib. No two native libs with the same name can be in the same jvm. We need to solve that for sure. Thanks On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi mailto:turcsa...@apache.org>> wrote: Hi Hai Luo, @RequiresInstanceClassLoading makes possible to configure separate / isolated "Additional Classpath Resources" settings on your HDFS processors (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). Is there any specific reason / use case why you are considering to remove it? Regards, Peter Turcsanyi On Mon, Nov 11, 2019 at 3:30 PM abellnotring mailto:abellnotr...@sina.com>> wrote: Hi,all I’m considering removing the RequiresInstanceClassLoading annotation from class AbstractHadoopProcessor, Does anybody know the potential Influence? Thanks By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
Hi,jeff, There is no Kerberos authentication in my HADOOP clusters , but I find UGI is initialized with an ExtendConfiguration(extend from hadoop Configruation) when those processor's instance was first scheduled .I would use nifi to connect different HADOOP cluster, will it run into any issues?(I’m running tests for this) Thanks By Hai Luo On 11/13/2019 07:01,Jeff wrote: If you remove the @RequiresInstanceClassloading, the UserGroupInformation class from Hadoop (hadoop-common, if I remember correctly) will be shared across all instances that come from a particular NAR (such as PutHDFS, ListHDFS, FetchHDFS, etc, from nifi-hadoop-nar-x.y.z.nar). If you are using Kerberos in those processors and configured different principals across the various processors, you could run into issues when the processors attempt to acquire new TGTs, most likely the first time a relogin is attempted. UGI has some static state and @RequiresInstanceClassloading makes sure each instance of a processor with that annotation has its own classloader to keep that kind of state from being shared across instances.On Mon, Nov 11, 2019 at 9:41 PM abellnotring <abellnotr...@sina.com> wrote: Hi,Peter & All I’m using kylo to manage the nifi flow(called feed in kylo),and there are 4200 instances(600+ instances extended from AbstractHadoopProcessor) in my nifi canvas,The NIFI Non-Heap Memory has increased more than 6GB after some days running ,which is extremely abnormal . I have analyzed the class loaded into Compressed Class Space ,and found most of the CCS was used by classes related by AbstractHadoopProcessor. So I think removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor may be a Solution for reducing the CCS used. Do you have any ideas for this ? Thanks By Hai Luo On 11/12/2019 02:17,Shawn Weeks wrote: I’m assuming your talking about the snappy problem. If you use compress content prior to puthdfs you can compress with Snappy as it uses the Java Native Snappy Lib. The HDFS processors are limited to the actual Hadoop Libraries so they’d have to change from Native to get around this. I’m pretty sure we need instance loading to handle the other issues mentioned. Thanks Shawn From: Joe Witt <joe.w...@gmail.com> Reply-To: "users@nifi.apache.org" <users@nifi.apache.org> Date: Monday, November 11, 2019 at 8:56 AM To: "users@nifi.apache.org" <users@nifi.apache.org> Subject: Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor Peter The most common challenge is if two isolated instances both want to use a native lib. No two native libs with the same name can be in the same jvm. We need to solve that for sure. Thanks On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi <turcsa...@apache.org> wrote: Hi Hai Luo, @RequiresInstanceClassLoading makes possible to configure separate / isolated "Additional Classpath Resources" settings on your HDFS processors (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). Is there any specific reason / use case why you are considering to remove it? Regards, Peter Turcsanyi On Mon, Nov 11, 2019 at 3:30 PM abellnotring <abellnotr...@sina.com> wrote: Hi,all I’m considering removing the RequiresInstanceClassLoading annotation from class AbstractHadoopProcessor, Does anybody know the potential Influence? Thanks By Hai Luo
Re: Influence about removing RequiresInstanceClassLoading from AbstractHadoopProcessor processor
By Hai Luo, I haven't personally attempted to use NiFi with Hadoop in the way you described. If your Hadoop clusters are not secured I don't think you'll have issues with UGI. I wouldn't recommend running NiFi and Hadoop clusters unsecured, however. On Tue, Nov 12, 2019 at 8:40 PM abellnotring wrote: > Hi,jeff, > There is no Kerberos authentication in my HADOOP clusters , but I > find UGI is initialized with an ExtendConfiguration(extend from hadoop > Configruation) when those processor's instance was first scheduled .I would > use nifi to connect different HADOOP cluster, will it run into any > issues?(I’m running tests for this) > > > Thanks > By Hai Luo > On 11/13/2019 07:01,Jeff wrote: > > If you remove the @RequiresInstanceClassloading, the UserGroupInformation > class from Hadoop (hadoop-common, if I remember correctly) will be shared > across all instances that come from a particular NAR (such as PutHDFS, > ListHDFS, FetchHDFS, etc, from nifi-hadoop-nar-x.y.z.nar). If you are > using Kerberos in those processors and configured different principals > across the various processors, you could run into issues when the > processors attempt to acquire new TGTs, most likely the first time a > relogin is attempted. UGI has some static state and > @RequiresInstanceClassloading makes sure each instance of a processor with > that annotation has its own classloader to keep that kind of state from > being shared across instances. > > On Mon, Nov 11, 2019 at 9:41 PM abellnotring > wrote: > >> Hi,Peter & All >> I’m using kylo to manage the nifi flow(called feed in kylo),and >> there are 4200 instances(600+ instances extended from >> AbstractHadoopProcessor) in my nifi canvas,The NIFI Non-Heap Memory has >> increased more than 6GB after some days running ,which is extremely >> abnormal . I have analyzed the class loaded into Compressed Class Space >> ,and found most of the CCS was used by classes related by >> AbstractHadoopProcessor. >>So I think removing RequiresInstanceClassLoading from >> AbstractHadoopProcessor processor may be a Solution for reducing the CCS >> used. >>Do you have any ideas for this ? >> >> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> >> >> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> >> >> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> >> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=abellnotring&uid=abellnotring%40sina.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22abellnotring%40sina.com%22%5D> >> Thanks >> >> >> By Hai Luo >> On 11/12/2019 02:17,Shawn Weeks >> wrote: >> >> I’m assuming your talking about the snappy problem. If you use compress >> content prior to puthdfs you can compress with Snappy as it uses the Java >> Native Snappy Lib. The HDFS processors are limited to the actual Hadoop >> Libraries so they’d have to change from Native to get around this. I’m >> pretty sure we need instance loading to handle the other issues mentioned. >> >> >> >> Thanks >> >> Shawn >> >> >> >> *From: *Joe Witt >> *Reply-To: *"users@nifi.apache.org" >> *Date: *Monday, November 11, 2019 at 8:56 AM >> *To: *"users@nifi.apache.org" >> *Subject: *Re: Influence about removing RequiresInstanceClassLoading >> from AbstractHadoopProcessor processor >> >> >> >> Peter >> >> >> >> The most common challenge is if two isolated instances both want to use a >> native lib. No two native libs with the same name can be in the same jvm. >> We need to solve that for sure. >> >> >> >> Thanks >> >> >> >> On Mon, Nov 11, 2019 at 9:53 AM Peter Turcsanyi >> wrote: >> >> Hi Hai Luo, >> >> >> >> @RequiresInstanceClassLoading makes possible to configure separate / >> isolated "Additional Classpath Resources" settings on your HDFS processors >> (eg. S3 storage driver on one of your PutHDFS and Azure Blob on the other). >> >> >> >> Is there any specific reason / use case why you are considering to remove >> it? >> >> >> >> Regards, >> >> Peter Turcsanyi >> >> >> >> On Mon, Nov 11, 2019 at 3:30 PM abellnotring >> wrote: >> >> Hi,all >> >> I’m considering removing the RequiresInstanceClassLoading annotation >> from class AbstractHadoopProcessor, >> >> Does anybody know the potential Influence? >> >> >> >> Thanks >> >> By Hai Luo >> >>